bug fix missing manifest

bug fix double sys import
update pdoc
2026-05-18 16:57:02 -03:00 · 2026-05-18 15:46:21 -03:00 · 2026-05-18 14:11:28 -03:00 · 2026-05-18 14:07:24 -03:00 · 2026-05-15 23:28:22 -03:00 · 2026-05-15 18:24:48 -03:00
97 changed files with 9491 additions and 2691 deletions
@@ -0,0 +1,22 @@
 .git
 __pycache__
 *.pyc
 *.pyo
 *.pyd
 .pytest_cache
 .venv
 venv
 env
 node_modules
 dist
 build
 *.egg-info
 docker
 docker-compose.yml
 .gemini
 .github
 docs
 scratch
 testall
 testremote
 automation-template.yaml
@@ -50,6 +50,7 @@ coverage.xml
 *.py,cover
 .hypothesis/
 .pytest_cache/
 scratch/
 # Translations
 *.mo
@@ -161,3 +162,11 @@ async_interact_plan.md
 repo_consolidado_limpio.md
 connpy_roadmap.md
 MULTI_USER_PLAN.md
 COPILOT_PLAN.md
 ARCHITECTURAL_DEBT_REFACTOR.md
 COPILOT_UI_FEATURES.md
 MULTI_USER_IMPLEMENTATION_STEPS.md
 #themes
 nord.yml
 theme.py
@@ -0,0 +1,8 @@
 include LICENSE
 include README.md
 include requirements.txt
 recursive-include connpy/core_plugins *
 recursive-include connpy/proto *
 recursive-include connpy/grpc_layer *
 recursive-exclude * __pycache__
 recursive-exclude * *.py[co]
@@ -9,519 +9,182 @@
 [![](https://img.shields.io/pypi/l/connpy.svg?style=flat-square)](https://github.com/fluzzi/connpy/blob/main/LICENSE)
 [![](https://img.shields.io/pypi/dm/connpy.svg?style=flat-square)](https://pypi.org/pypi/connpy/)
-Connpy is a SSH, SFTP, Telnet, kubectl, Docker pod, and AWS SSM connection manager and automation module for Linux, Mac, and Docker.
+**Connpy** is a powerful Connection Manager and Network Automation Platform for Linux, Mac, and Docker. It provides a unified interface for **SSH, SFTP, Telnet, kubectl, Docker pods, and AWS SSM**.
 The v6 release introduces the **AI Copilot**, an interactive terminal assistant that understands your network context and helps you manage your infrastructure more intelligently.
 ## 🤖 AI Copilot (New in v6)
 The AI Copilot is deeply integrated into your terminal workflow:
 - **Terminal Context Awareness**: The Copilot can "see" your screen output, helping you diagnose errors or analyze command results in real-time.
 - **Hybrid Multi-Agent System**: Automatically escalates complex tasks between the **Network Engineer** (execution) and the **Network Architect** (strategy).
 - **MCP Integration**: Dynamically load tools from external providers (6WIND, AWS, etc.) via the Model Context Protocol.
 - **Interactive Chat**: Launch with `conn ai` for a collaborative troubleshooting session.
 ## Core Features
 - **Multi-Protocol**: Native support for SSH, SFTP, Telnet, kubectl, Docker exec, and AWS SSM.
 - **Context Management**: Set regex-based contexts to manage specific nodes across different environments (work, home, clients).
 - **Advanced Inventory**:
    - Organize nodes in folders (`@folder`) and subfolders (`@subfolder@folder`).
    - Use Global Profiles (`@profilename`) to manage shared credentials easily.
    - Bulk creation, copying, moving, and export/import of nodes.
 - **Modern UI**: High-performance terminal experience with `prompt-toolkit`, including:
    - Fuzzy search integration with `fzf`.
    - Advanced tab completion.
    - Syntax highlighting and customizable themes.
 - **Automation Engine**: Run parallel tasks and playbooks on multiple devices with variable support.
 - **Plugin System**: Build and execute custom Python scripts locally or on a remote gRPC server.
 - **gRPC Architecture**: Fully decoupled Client/Server model for distributed management.
 - **Privacy & Sync**: Local-first encrypted storage (RSA/OAEP) with optional Google Drive backup.
 ## Installation
 ```bash
 pip install connpy
 ### Run it in Windows using docker
 ```
 ### Run it in Windows/Linux using Docker
 ```bash
 git clone https://github.com/fluzzi/connpy
-docker compose -f path/to/folder/docker-compose.yml build
+cd connpy
-docker compose -f path/to/folder/docker-compose.yml run -it connpy-app
+docker compose build
 # Run it like a native app (completely silent)
 docker compose run --rm --remove-orphans connpy-app [command]
 # Pro Tip: Add this alias for a 100% native experience from any folder
 alias conn='docker compose -f /path/to/connpy/docker-compose.yml run --rm --remove-orphans connpy-app'
 ```
-## Connection manager 
+---
 ## 🔒 Privacy & Integration
 ### Privacy Policy
-
+Connpy is committed to protecting your privacy:
-Connpy is committed to protecting your privacy. Our privacy policy explains how we handle user data:
+- **Local Storage**: All server addresses, usernames, and passwords are encrypted and stored **only** on your machine. No data is transmitted to our servers.
-
+- **Data Access**: Data is used solely for managing and automating your connections.
 - **Data Access**: Connpy accesses data necessary for managing remote host connections, including server addresses, usernames, and passwords. This data is stored locally on your machine and is not transmitted or shared with any third parties.
 - **Data Usage**: User data is used solely for the purpose of managing and automating SSH, Telnet, and SSM connections.
 - **Data Storage**: All connection details are stored locally and securely on your device. We do not store or process this data on our servers.
 - **Data Sharing**: We do not share any user data with third parties.
 ### Google Integration
 Used strictly for backup:
 - **Backup**: Sync your encrypted configuration with your Google Drive account.
 - **Scoped Access**: Connpy only accesses its own backup files.
-Connpy integrates with Google services for backup purposes:
+---
- **Configuration Backup**: The app allows users to store their device information in the app configuration. This configuration can be synced with Google services to create backups.
+## Usage
 - **Data Access**: Connpy only accesses its own files and does not access any other files on your Google account.
 - **Data Usage**: The data is used solely for backup and restore purposes, ensuring that your device information and configurations are safe and recoverable.
 - **Data Sharing**: Connpy does not share any user data with third parties, including Google. The backup data is only accessible by the user.
-For more detailed information, please read our [Privacy Policy](https://connpy.gederico.dynu.net/fluzzi32/connpy/src/branch/main/PRIVATE_POLICY.md).
+```bash
 ### Features
    - Manage connections using SSH, SFTP, Telnet, kubectl, Docker exec, and AWS SSM.
    - Set contexts to manage specific nodes from specific contexts (work/home/clients/etc).
    - You can generate profiles and reference them from nodes using @profilename so you don't
      need to edit multiple nodes when changing passwords or other information.
    - Nodes can be stored on @folder or @subfolder@folder to organize your devices. They can
      be referenced using node@subfolder@folder or node@folder.
    - If you have too many nodes, get a completion script using: conn config --completion.
      Or use fzf by installing pyfzf and running conn config --fzf true.
    - Create in bulk, copy, move, export, and import nodes for easy management.
    - Run automation scripts on network devices.
    - Use AI with a multi-agent system (Engineer/Architect) to manage devices.
      Supports any LLM provider via litellm (OpenAI, Anthropic, Google, etc.).
      Features streaming responses, interactive chat, and extensible plugin tools.
    - Add plugins with your own scripts, and execute them remotely.
    - Fully decoupled gRPC Client/Server architecture.
    - Unified UI with syntax highlighting and theming.
    - Much more!
 ### Usage:
 ```
 usage: conn [-h] [--add | --del | --mod | --show | --debug] [node|folder] [--sftp]
-       conn {profile,move,mv,copy,cp,list,ls,bulk,export,import,ai,run,api,plugin,config,sync,context} ...
+       conn {profile,move,copy,list,bulk,export,import,ai,run,api,plugin,config,sync,context} ...
 positional arguments:
  node|folder        node[@subfolder][@folder]
                     Connect to specific node or show all matching nodes
                     [@subfolder][@folder]
                     Show all available connections globally or in specified path
 options:
  -h, --help         show this help message and exit
  -v, --version      Show version
  -a, --add          Add new node[@subfolder][@folder] or [@subfolder]@folder
  -r, --del, --rm    Delete node[@subfolder][@folder] or [@subfolder]@folder
  -e, --mod, --edit  Modify node[@subfolder][@folder]
  -s, --show         Show node[@subfolder][@folder]
  -d, --debug        Display all conections steps
  -t, --sftp         Connects using sftp instead of ssh
  --service-mode     Set the backend service mode (local or remote)
  --remote           Connect to a remote connpy service via gRPC
  --theme            UI Output theme (dark, light, or path)
 Commands:
  profile         Manage profiles
  move(mv)        Move node
  copy(cp)        Copy node
  list(ls)        List profiles, nodes or folders
  bulk            Add nodes in bulk
  export          Export connection folder to Yaml file
  import          Import connection folder to config from Yaml file
  ai              Make request to an AI
  run             Run scripts or commands on nodes
  api             Start and stop connpy api
  plugin          Manage plugins
  config          Manage app config
  sync            Sync config with Google
  context         Manage contexts with regex matching
 ```
-### Manage profiles:
+### Basic Examples:
-```
+```bash
-usage: conn profile [-h] (--add | --del | --mod | --show) profile
+# Add a folder and subfolder
 conn --add @office
 conn --add @datacenter@office
-positional arguments:
+# Add a node with a profile
-  profile        Name of profile to manage
+conn --add server1@datacenter@office --profile @myuser
-options:
+# Connect to a node (fuzzy match)
-  -h, --help         show this help message and exit
+conn server1
  -a, --add          Add new profile
  -r, --del, --rm    Delete profile
  -e, --mod, --edit  Modify profile
  -s, --show         Show profile
 # Start the AI Copilot
 conn ai
 # Run a command on all nodes in a folder
 conn run @office "uptime"
 ```
-### Examples:
+---
-```
+
-   #Add new profile
+## 🔌 Plugin System
-   conn profile --add office-user
+Connpy supports a robust plugin architecture where scripts can run transparently on a remote gRPC server.
-   #Add new folder
+
-   conn --add @office
+### Structure
-   #Add new subfolder
+Plugins must be Python files containing:
-   conn --add @datacenter@office
+- **Class `Parser`**: Defines `argparse` arguments.
-   #Add node to subfolder
+- **Class `Entrypoint`**: Execution logic.
-   conn --add server@datacenter@office
+- **Class `Preload`**: (Optional) Hooks and modifications to the core app.
-   #Add node to folder
+
-   conn --add pc@office
+See the [Plugin Requirements section](#plugin-requirements-for-connpy) for full technical details.
-   #Show node information
+
-   conn --show server@datacenter@office
+---
-   #Connect to nodes
+
   conn pc@office
   conn server
   #Create and set new context
   conn context -a office .*@office
   conn context --set office
   #Run a command in a node
   conn run server ls -la
 ``` 
 ## Plugin Requirements for Connpy
 ### Remote Plugin Execution
 When Connpy operates in remote mode, plugins are executed **transparently on the server**:
 - The client automatically downloads the plugin source code (`Parser` class context) to generate the local `argparse` structure and provide autocompletion.
- The execution phase (`Entrypoint` class) is redirected via gRPC streams to execute in the server's memory, ensuring the plugin runs securely against the server's inventory without passing sensitive data to the client.
+- The execution phase (`Entrypoint` class) is redirected via gRPC streams to execute in the server's memory.
- You can manage remote plugins using the `--remote` flag (e.g. `connpy plugin --add myplugin script.py --remote`).
+- You can manage remote plugins using the `--remote` flag.
 ### General Structure
- The plugin script must be a Python file.
+- The plugin script must define specific classes:
- Only the following top-level elements are allowed in the plugin script:
+  1. **Class `Parser`**: Handles `argparse.ArgumentParser` initialization.
-  - Class definitions
+  2. **Class `Entrypoint`**: Main execution logic (receives `args`, `parser`, and `connapp`).
-  - Function definitions
+  3. **Class `Preload`**: (Optional) For modifying core app behavior or registering hooks.
  - Import statements
  - The `if __name__ == "__main__":` block for standalone execution
  - Pass statements
 ### Specific Class Requirements
 - The plugin script must define specific classes with particular attributes and methods. Each class serves a distinct role within the plugin's architecture:
  1. **Class `Parser`**:
     - **Purpose**: Handles parsing of command-line arguments.
     - **Requirements**:
       - Must contain only one method: `__init__`.
       - The `__init__` method must initialize at least one attribute:
         - `self.parser`: An instance of `argparse.ArgumentParser`.
  2. **Class `Entrypoint`**:
     - **Purpose**: Acts as the entry point for plugin execution, utilizing parsed arguments and integrating with the main application.
     - **Requirements**:
       - Must have an `__init__` method that accepts exactly three parameters besides `self`:
         - `args`: Arguments passed to the plugin.
         - The parser instance (typically `self.parser` from the `Parser` class).
         - The Connapp instance to interact with the Connpy app.
  3. **Class `Preload`**:
     - **Purpose**: Performs any necessary preliminary setup or configuration independent of the main parsing and entry logic.
   - **Requirements**:
     - Contains at least an `__init__` method that accepts parameter connapp besides `self`.
 ### Class Dependencies and Combinations
 - **Dependencies**:
  - `Parser` and `Entrypoint` are interdependent and must both be present if one is included.
  - `Preload` is independent and may exist alone or alongside the other classes.
 - **Valid Combinations**:
  - `Parser` and `Entrypoint` together.
  - `Preload` alone.
  - All three classes (`Parser`, `Entrypoint`, `Preload`).
 ### Preload Modifications and Hooks
-
+You can customize the behavior of core classes using hooks:
-In the `Preload` class of the plugin system, you have the ability to customize the behavior of existing classes and methods within the application through a robust hooking system. This documentation explains how to use the `modify`, `register_pre_hook`, and `register_post_hook` methods to tailor plugin functionality to your needs.
+- **`modify(method)`**: Alter class instances (e.g., `connapp.config`, `connapp.ai`).
-
+- **`register_pre_hook(method)`**: Logic to run before a method execution.
-#### Modifying Classes with `modify`
+- **`register_post_hook(method)`**: Logic to run after a method execution.
 The `modify` method allows you to alter instances of a class at the time they are created or after their creation. This is particularly useful for setting or modifying configuration settings, altering default behaviors, or adding new functionalities to existing classes without changing the original class definitions.
 - **Usage**: Modify a class to include additional configurations or changes
 - **Modify Method Signature**:
  - `modify(modification_method)`: A function that is invoked with an instance of the class as its argument. This function should perform any modifications directly on this instance.
 - **Modification Method Signature**:
  - **Arguments**:
    - `cls`:  This function accepts a single argument, the class instance, which it then modifies.
  - **Modifiable Classes**:
    - `connapp.config`
    - `connapp.node`
    - `connapp.nodes`
    - `connapp.ai`
  - ```python
    def modify_config(cls):
        # Example modification: adding a new attribute or modifying an existing one
        cls.new_attribute = 'New Value'
    class Preload:
        def __init__(self, connapp):
            # Applying modification to the config class instance
            connapp.config.modify(modify_config)
    ```
 #### Implementing Method Hooks
 There are 2 methods that allows you to define custom logic to be executed before (`register_pre_hook`) or after (`register_post_hook`) the main logic of a method. This is particularly useful for logging, auditing, preprocessing inputs, postprocessing outputs or adding functionalities.
  - **Usage**: Register hooks to methods to execute additional logic before or after the main method execution.
 - **Registration Methods Signature**:
  - `register_pre_hook(pre_hook_method)`: A function that is invoked before the main method is executed. This function should do preprocessing of the arguments.
  - `register_post_hook(post_hook_method)`: A function that is invoked after the main method is executed. This function should do postprocessing of the outputs.
 - **Method Signatures for Pre-Hooks**
  - `pre_hook_method(*args, **kwargs)`
  - **Arguments**:
    - `*args`, `**kwargs`: The arguments and keyword arguments that will be passed to the method being hooked. The pre-hook function has the opportunity to inspect and modify these arguments before they are passed to the main method.
  - **Return**:
    - Must return a tuple `(args, kwargs)`, which will be used as the new arguments for the main method. If the original arguments are not modified, the function should return them as received.
 - **Method Signatures for Post-Hooks**:
  - `post_hook_method(*args, **kwargs)`
  - **Arguments**:
    - `*args`, `**kwargs`: The arguments and keyword arguments that were passed to the main method.
      - `kwargs["result"]`: The value returned by the main method. This allows the post-hook to inspect and even alter the result before it is returned to the original caller.
  - **Return**:
    - Can return a modified result, which will replace the original result of the main method, or simply return `kwargs["result"]` to return the original method result.    
  - ```python
    def pre_processing_hook(*args, **kwargs):
        print("Pre-processing logic here")
        # Modify arguments or perform any checks
        return args, kwargs  # Return modified or unmodified args and kwargs
    def post_processing_hook(*args, **kwargs):
        print("Post-processing logic here")
        # Modify the result or perform any final logging or cleanup
        return kwargs["result"]  # Return the modified or unmodified result
    class Preload:
        def __init__(self, connapp):
            # Registering a pre-hook
            connapp.ai.some_method.register_pre_hook(pre_processing_hook)
            # Registering a post-hook
            connapp.node.another_method.register_post_hook(post_processing_hook)
    ```
 ### Executable Block
 - The plugin script can include an executable block:
  - `if __name__ == "__main__":`
  - This block allows the plugin to be run as a standalone script for testing or independent use.
 ### Command Completion Support
 Plugins can provide intelligent tab completion:
 1. **Tree-based Completion (Recommended)**: Define `_connpy_tree(info)` returning a navigation dictionary.
 2. **Legacy Completion**: Define `_connpy_completion(wordsnumber, words, info)`.
-Plugins can provide intelligent **tab completion** by defining autocompletion logic. There are two supported methods, with the tree-based approach being the most modern and recommended.
+---
-#### 1. Tree-based Completion (Recommended)
+## ⚙️ gRPC Service Architecture
 Connpy can operate in a decoupled mode:
 1. **Start the API (Server)**: `conn api -s 50051`
 2. **Configure the Client**:
   ```bash
   conn config --service-mode remote
   conn config --remote-host localhost:50051
   ```
 All inventory management and execution will now happen on the server.
-Define a function called `_connpy_tree` that returns a declarative navigation tree. This method is highly efficient, supports complex state loops, and is very simple to implement for most use cases.
+---
 ## 🐍 Automation Module (API)
 You can use `connpy` as a Python library for your own scripts.
 ### Basic Execution
 ```python
 def _connpy_tree(info=None):
    nodes = info.get("nodes", [])
    return {
        "__exclude_used__": True,  # Filter out words already typed
        "__extra__": nodes,        # Suggest nodes at this level
        "--format": ["json", "yaml", "table"], # Fixed suggestions
        "*": {                     # Wildcard matches any positional word
            "interface1": None,
            "interface2": None,
            "--verbose": None
        }
    }
 ```
 - **Keys**: Literal completions (exact matches).
 - **`*` Key**: A wildcard that matches any positional word typed by the user.
 - **`__extra__`**: A list or a callable `(words) -> list` that adds dynamic suggestions.
 - **`__exclude_used__`**: (Boolean) If True, automatically filters out words already present in the command line.
 #### 2. Legacy Function-based Completion
 For backward compatibility or highly custom logic, you can define `_connpy_completion`.
 ```python
 def _connpy_completion(wordsnumber, words, info=None):
    if wordsnumber == 3:
        return ["--help", "--verbose", "start", "stop"]
    elif wordsnumber == 4 and words[2] == "start":
        return info["nodes"]  # Suggest node names
    return []
 ```
 | Parameter      | Description |
 |----------------|-------------|
 | `wordsnumber`  | Integer indicating the total number of words on the command line. For plugins, this typically starts at 3. |
 | `words`        | A list of tokens (words) already typed. `words[0]` is always the name of the plugin. |
 | `info`         | A dictionary of structured context data (`nodes`, `folders`, `profiles`, `config`). |
 > In this example, if the user types `connpy myplugin start ` and presses Tab, it will suggest node names.
 ### Handling Unknown Arguments
 Plugins can choose to accept and process unknown arguments that are **not explicitly defined** in the parser. To enable this behavior, the plugin must define the following hidden argument in its `Parser` class:
 ```
 self.parser.add_argument(
    "--unknown-args",
    action="store_true",
    default=True,
    help=argparse.SUPPRESS
 )
 ```
 #### Behavior:
 - When this argument is present, Connpy will parse the known arguments and capture any extra (unknown) ones.
 - These unknown arguments will be passed to the plugin as `args.unknown_args` inside the `Entrypoint`.
 - If the user does not pass any unknown arguments, `args.unknown_args` will contain the default value (`True`, unless overridden).
 #### Example:
 If a plugin accepts unknown tcpdump flags like this:
 ```
 connpy myplugin -nn -s0
 ```
 And defines the hidden `--unknown-args` flag as shown above, then:
 - `args.unknown_args` inside `Entrypoint.__init__()` will be: `['-nn', '-s0']`
 > This allows the plugin to receive and process arguments intended for external tools (e.g., `tcpdump`) without argparse raising an error.
 #### Note:
 If a plugin does **not** define `--unknown-args`, any extra arguments passed will cause argparse to fail with an unrecognized arguments error.
 ### Script Verification
 - The `verify_script` method in `plugins.py` is used to check the plugin script's compliance with these standards.
 - Non-compliant scripts will be rejected to ensure consistency and proper functionality within the plugin system.
 ### Example Script
 For a practical example of how to write a compatible plugin script, please refer to the following example:
 [Example Plugin Script](https://github.com/fluzzi/awspy)
 This script demonstrates the required structure and implementation details according to the plugin system's standards.
 ## Automation module usage
 ### Standalone module
 ```
 import connpy
-router = connpy.node("uniqueName","ip/host", user="username", password="password")
+router = connpy.node("uniqueName", "1.1.1.1", user="admin")
-router.run(["term len 0","show run"])
+router.run(["show ip int brief"])
 print(router.output)
 hasip = router.test("show ip int brief","1.1.1.1")
 if hasip:
    print("Router has ip 1.1.1.1")
 else:
    print("router does not have ip 1.1.1.1")
 ```
-### Using manager configuration
+### Parallel Tasks with Variables
-```
+```python
 import connpy
 conf = connpy.configfile()
 device = conf.getitem("router@office")
 router = connpy.node("unique name", **device, config=conf)
 result = router.run("show ip int brief")
 print(result)
 ```
 ### Running parallel tasks on multiple devices 
 ```
 import connpy
 conf = connpy.configfile()
 #You can get the nodes from the config from a folder and fitlering in it
 nodes = conf.getitem("@office", ["router1", "router2", "router3"])
 #You can also get each node individually:
 nodes = {}
 nodes["router1"] = conf.getitem("router1@office")
 nodes["router2"] = conf.getitem("router2@office")
 nodes["router10"] = conf.getitem("router10@datacenter")
 #Also, you can create the nodes manually:
 nodes = {}
 nodes["router1"] = {"host": "1.1.1.1", "user": "user", "password": "password1"}
 nodes["router2"] = {"host": "1.1.1.2", "user": "user", "password": "password2"}
 nodes["router3"] = {"host": "1.1.1.2", "user": "user", "password": "password3"}
 #Finally you run some tasks on the nodes
 mynodes = connpy.nodes(nodes, config = conf)
 result = mynodes.test(["show ip int br"], "1.1.1.2")
 for i in result:
    print("---" + i + "---")
    print(result[i])
    print()
 # Or for one specific node
 mynodes.router1.run(["term len 0". "show run"], folder = "/home/user/logs")
 ```
 ### Using variables
 ```
 import connpy
 config = connpy.configfile()
-nodes = config.getitem("@office", ["router1", "router2", "router3"])
+nodes = config.getitem("@office", ["router1", "router2"])
-commands = []
+routers = connpy.nodes(nodes, config=config)
-commands.append("config t")
+
-commands.append("interface lo {id}")
+variables = {
-commands.append("ip add {ip} {mask}")
+    "router1@office": {"id": "1"},
-commands.append("end")
+    "__global__": {"mask": "255.255.255.0"}
-variables = {}
+}
-variables["router1@office"] = {"ip": "10.57.57.1"}
+routers.run(["interface lo{id}", "ip address 10.0.0.{id} {mask}"], variables)
 variables["router2@office"] = {"ip": "10.57.57.2"}
 variables["router3@office"] = {"ip": "10.57.57.3"}
 variables["__global__"] = {"id": "57"}
 variables["__global__"]["mask"] =  "255.255.255.255"
 expected = "!"
 routers = connpy.nodes(nodes, config = config)
 routers.run(commands, variables)
 routers.test("ping {ip}", expected, variables)
 for key in routers.result:
    print(key, ' ---> ', ("pass" if routers.result[key] else "fail"))
 ```
-### Using AI
+
-The AI module uses a multi-agent architecture with an **Engineer** (fast execution) and an **Architect** (strategic reasoning). It supports any LLM provider through [litellm](https://github.com/BerriAI/litellm).
+### AI Programmatic Use
 ```python
 import connpy
-conf = connpy.configfile()
+myai = connpy.ai(connpy.configfile())
-# Uses models and API keys from config, or override them:
+response = myai.ask("What is the status of the BGP neighbors in the office?")
 myai = connpy.ai(conf, engineer_model="gemini/gemini-2.5-flash", engineer_api_key="your-key")
 result = myai.ask("go to router1 and show me the running configuration")
 print(result["response"])
 # Streaming is enabled by default for CLI, disable for programmatic use:
 result = myai.ask("show interfaces on all routers", stream=False)
 print(result["response"])
 ```
-#### AI Plugin Tool Registration
+---
-Plugins can extend the AI system by registering custom tools via the `Preload` class:
+*For detailed developer notes and plugin hooks documentation, see the [Documentation](https://fluzzi.github.io/connpy/).*
 ```python
 def _register_my_tools(ai_instance):
    tool_def = {
        "type": "function",
        "function": {
            "name": "my_custom_tool",
            "description": "Does something useful.",
            "parameters": {
                "type": "object",
                "properties": {"query": {"type": "string"}},
                "required": ["query"]
            }
        }
    }
    ai_instance.register_ai_tool(
        tool_definition=tool_def,
        handler=my_handler_function,
        target="engineer",  # or "architect" or "both"
        engineer_prompt="- My tool: does X.",
        architect_prompt="  * My tool (my_custom_tool)."
    )
 class Preload:
    def __init__(self, connapp):
        connapp.ai.modify(_register_my_tools)
 ```
 ## gRPC Service Architecture
 Connpy features a completely decoupled gRPC Client/Server architecture. You can run Connpy as a standalone background service and connect to it remotely via the CLI or other clients.
 ### 1. Start the Server
 Start the gRPC service by running:
 ```bash
 connpy api -s 50051
 ```
 The server will handle all configurations, connections, AI sessions, and plugin execution locally on the machine it runs on.
 ### 2. Connect the Client
 Configure your local CLI client to connect to the remote server:
 ```bash
 connpy config --service-mode remote
 connpy config --remote-host localhost:50051
 ```
 Once configured, all commands (`connpy node`, `connpy list`, `connpy ai`, etc.) will execute transparently on the remote server via thin-client proxies. You can revert back to standalone execution at any time by running `connpy config --service-mode local`.
 ### Programmatic Access (gRPC & SOA)
 If you wish to build your own application (Web, Desktop, or Scripts) using the Connpy backend, you can use the `ServiceProvider` to interact with either a local or remote service transparently.
 ```python
 import connpy
 from connpy.services.provider import ServiceProvider
 # Initialize local config
 config = connpy.configfile()
 # Connect to the remote gRPC service
 services = ServiceProvider(
    config, 
    mode="remote", 
    remote_host="localhost:50051"
 )
 # Use any service (the logic is identical to local mode)
 nodes = services.nodes.list_nodes()
 for name in nodes:
    print(f"Found node: {name}")
 # Run a command remotely via streaming
 for chunk in services.execution.run_commands(nodes=["server1"], commands=["uptime"]):
    print(chunk["output"], end="")
 ```
@@ -1,476 +1,182 @@
 #!/usr/bin/env python3
 '''
-## Connection manager
+<p align="center">
  <img src="https://nginx.gederico.dynu.net/images/CONNPY-resized.png" alt="App Logo">
 </p>
 Connpy is a SSH, SFTP, Telnet, kubectl, Docker pod, and AWS SSM connection manager and automation module for Linux, Mac, and Docker.
-### Features
+# Connpy
-    - Manage connections using SSH, SFTP, Telnet, kubectl, Docker exec, and AWS SSM.
+[![](https://img.shields.io/pypi/v/connpy.svg?style=flat-square)](https://pypi.org/pypi/connpy/)
-    - Set contexts to manage specific nodes from specific contexts (work/home/clients/etc).
+[![](https://img.shields.io/pypi/pyversions/connpy.svg?style=flat-square)](https://pypi.org/pypi/connpy/)
-    - You can generate profiles and reference them from nodes using @profilename so you don't
+[![](https://img.shields.io/pypi/l/connpy.svg?style=flat-square)](https://github.com/fluzzi/connpy/blob/main/LICENSE)
-      need to edit multiple nodes when changing passwords or other information.
+[![](https://img.shields.io/pypi/dm/connpy.svg?style=flat-square)](https://pypi.org/pypi/connpy/)
    - Nodes can be stored on @folder or @subfolder@folder to organize your devices. They can
      be referenced using node@subfolder@folder or node@folder.
    - If you have too many nodes, get a completion script using: conn config --completion.
      Or use fzf by installing pyfzf and running conn config --fzf true.
    - Create in bulk, copy, move, export, and import nodes for easy management.
    - Run automation scripts on network devices.
    - Use AI with a multi-agent system (Engineer/Architect) to help you manage your devices.
      Supports any LLM provider via litellm (OpenAI, Anthropic, Google, etc.).
    - Add plugins with your own scripts, and execute them remotely.
    - Fully decoupled gRPC Client/Server architecture.
    - Unified UI with syntax highlighting and theming.
    - Much more!
-### Usage
+**Connpy** is a powerful Connection Manager and Network Automation Platform for Linux, Mac, and Docker. It provides a unified interface for **SSH, SFTP, Telnet, kubectl, Docker pods, and AWS SSM**.
 The v6 release introduces the **AI Copilot**, an interactive terminal assistant that understands your network context and helps you manage your infrastructure more intelligently.
 ## 🤖 AI Copilot (New in v6)
 The AI Copilot is deeply integrated into your terminal workflow:
 - **Terminal Context Awareness**: The Copilot can "see" your screen output, helping you diagnose errors or analyze command results in real-time.
 - **Hybrid Multi-Agent System**: Automatically escalates complex tasks between the **Network Engineer** (execution) and the **Network Architect** (strategy).
 - **MCP Integration**: Dynamically load tools from external providers (6WIND, AWS, etc.) via the Model Context Protocol.
 - **Interactive Chat**: Launch with `conn ai` for a collaborative troubleshooting session.
 ## Core Features
 - **Multi-Protocol**: Native support for SSH, SFTP, Telnet, kubectl, Docker exec, and AWS SSM.
 - **Context Management**: Set regex-based contexts to manage specific nodes across different environments (work, home, clients).
 - **Advanced Inventory**:
    - Organize nodes in folders (`@folder`) and subfolders (`@subfolder@folder`).
    - Use Global Profiles (`@profilename`) to manage shared credentials easily.
    - Bulk creation, copying, moving, and export/import of nodes.
 - **Modern UI**: High-performance terminal experience with `prompt-toolkit`, including:
    - Fuzzy search integration with `fzf`.
    - Advanced tab completion.
    - Syntax highlighting and customizable themes.
 - **Automation Engine**: Run parallel tasks and playbooks on multiple devices with variable support.
 - **Plugin System**: Build and execute custom Python scripts locally or on a remote gRPC server.
 - **gRPC Architecture**: Fully decoupled Client/Server model for distributed management.
 - **Privacy & Sync**: Local-first encrypted storage (RSA/OAEP) with optional Google Drive backup.
 ## Installation
 ```bash
 pip install connpy
 ```
 ### Run it in Windows/Linux using Docker
 ```bash
 git clone https://github.com/fluzzi/connpy
 cd connpy
 docker compose build
 # Run it like a native app (completely silent)
 docker compose run --rm --remove-orphans connpy-app [command]
 # Pro Tip: Add this alias for a 100% native experience from any folder
 alias conn='docker compose -f /path/to/connpy/docker-compose.yml run --rm --remove-orphans connpy-app'
 ```
 ---
 ## 🔒 Privacy & Integration
 ### Privacy Policy
 Connpy is committed to protecting your privacy:
 - **Local Storage**: All server addresses, usernames, and passwords are encrypted and stored **only** on your machine. No data is transmitted to our servers.
 - **Data Access**: Data is used solely for managing and automating your connections.
 ### Google Integration
 Used strictly for backup:
 - **Backup**: Sync your encrypted configuration with your Google Drive account.
 - **Scoped Access**: Connpy only accesses its own backup files.
 ---
 ## Usage
 ```bash
 usage: conn [-h] [--add | --del | --mod | --show | --debug] [node|folder] [--sftp]
-       conn {profile,move,mv,copy,cp,list,ls,bulk,export,import,ai,run,api,plugin,config,sync,context} ...
+       conn {profile,move,copy,list,bulk,export,import,ai,run,api,plugin,config,sync,context} ...
 positional arguments:
  node|folder        node[@subfolder][@folder]
                     Connect to specific node or show all matching nodes
                     [@subfolder][@folder]
                     Show all available connections globally or in specified path
 options:
  -h, --help         show this help message and exit
  -v, --version      Show version
  -a, --add          Add new node[@subfolder][@folder] or [@subfolder]@folder
  -r, --del, --rm    Delete node[@subfolder][@folder] or [@subfolder]@folder
  -e, --mod, --edit  Modify node[@subfolder][@folder]
  -s, --show         Show node[@subfolder][@folder]
  -d, --debug        Display all conections steps
  -t, --sftp         Connects using sftp instead of ssh
  --service-mode     Set the backend service mode (local or remote)
  --remote           Connect to a remote connpy service via gRPC
  --theme            UI Output theme (dark, light, or path)
 Commands:
  profile         Manage profiles
  move(mv)        Move node
  copy(cp)        Copy node
  list(ls)        List profiles, nodes or folders
  bulk            Add nodes in bulk
  export          Export connection folder to Yaml file
  import          Import connection folder to config from Yaml file
  ai              Make request to an AI
  run             Run scripts or commands on nodes
  api             Start and stop connpy api
  plugin          Manage plugins
  config          Manage app config
  sync            Sync config with Google
  context         Manage contexts with regex matching
 ```
-###   Manage profiles
+### Basic Examples:
-```
+```bash
-usage: conn profile [-h] (--add | --del | --mod | --show) profile
+# Add a folder and subfolder
 conn --add @office
 conn --add @datacenter@office
-positional arguments:
+# Add a node with a profile
-  profile        Name of profile to manage
+conn --add server1@datacenter@office --profile @myuser
-options:
+# Connect to a node (fuzzy match)
-  -h, --help         show this help message and exit
+conn server1
  -a, --add          Add new profile
  -r, --del, --rm    Delete profile
  -e, --mod, --edit  Modify profile
  -s, --show         Show profile
 # Start the AI Copilot
 conn ai
 # Run a command on all nodes in a folder
 conn run @office "uptime"
 ```
-###   Examples
+---
-```
+
   #Add new profile
   conn profile --add office-user
   #Add new folder
   conn --add @office
   #Add new subfolder
   conn --add @datacenter@office
   #Add node to subfolder
   conn --add server@datacenter@office
   #Add node to folder
   conn --add pc@office
   #Show node information
   conn --show server@datacenter@office
   #Connect to nodes
   conn pc@office
   conn server
   #Create and set new context
   conn context -a office .*@office
   conn context --set office
   #Run a command in a node
   conn run server ls -la
 ``` 
 ## Plugin Requirements for Connpy
 ### Remote Plugin Execution
 When Connpy operates in remote mode, plugins are executed **transparently on the server**:
 - The client automatically downloads the plugin source code (`Parser` class context) to generate the local `argparse` structure and provide autocompletion.
- The execution phase (`Entrypoint` class) is redirected via gRPC streams to execute in the server's memory, ensuring the plugin runs securely against the server's inventory without passing sensitive data to the client.
+- The execution phase (`Entrypoint` class) is redirected via gRPC streams to execute in the server's memory.
- You can manage remote plugins using the `--remote` flag (e.g. `connpy plugin --add myplugin script.py --remote`).
+- You can manage remote plugins using the `--remote` flag.
 ### General Structure
- The plugin script must be a Python file.
+- The plugin script must define specific classes:
- Only the following top-level elements are allowed in the plugin script:
+  1. **Class `Parser`**: Handles `argparse.ArgumentParser` initialization.
-  - Class definitions
+  2. **Class `Entrypoint`**: Main execution logic (receives `args`, `parser`, and `connapp`).
-  - Function definitions
+  3. **Class `Preload`**: (Optional) For modifying core app behavior or registering hooks.
  - Import statements
  - The `if __name__ == "__main__":` block for standalone execution
  - Pass statements
 ### Specific Class Requirements
 - The plugin script must define specific classes with particular attributes and methods. Each class serves a distinct role within the plugin's architecture:
  1. **Class `Parser`**:
     - **Purpose**: Handles parsing of command-line arguments.
     - **Requirements**:
       - Must contain only one method: `__init__`.
       - The `__init__` method must initialize at least one attribute:
         - `self.parser`: An instance of `argparse.ArgumentParser`.
  2. **Class `Entrypoint`**:
     - **Purpose**: Acts as the entry point for plugin execution, utilizing parsed arguments and integrating with the main application.
     - **Requirements**:
       - Must have an `__init__` method that accepts exactly three parameters besides `self`:
         - `args`: Arguments passed to the plugin.
         - The parser instance (typically `self.parser` from the `Parser` class).
         - The Connapp instance to interact with the Connpy app.
  3. **Class `Preload`**:
     - **Purpose**: Performs any necessary preliminary setup or configuration independent of the main parsing and entry logic.
   - **Requirements**:
     - Contains at least an `__init__` method that accepts parameter connapp besides `self`.
 ### Class Dependencies and Combinations
 - **Dependencies**:
  - `Parser` and `Entrypoint` are interdependent and must both be present if one is included.
  - `Preload` is independent and may exist alone or alongside the other classes.
 - **Valid Combinations**:
  - `Parser` and `Entrypoint` together.
  - `Preload` alone.
  - All three classes (`Parser`, `Entrypoint`, `Preload`).
 ### Preload Modifications and Hooks
-
+You can customize the behavior of core classes using hooks:
-In the `Preload` class of the plugin system, you have the ability to customize the behavior of existing classes and methods within the application through a robust hooking system. This documentation explains how to use the `modify`, `register_pre_hook`, and `register_post_hook` methods to tailor plugin functionality to your needs.
+- **`modify(method)`**: Alter class instances (e.g., `connapp.config`, `connapp.ai`).
-
+- **`register_pre_hook(method)`**: Logic to run before a method execution.
-#### Modifying Classes with `modify`
+- **`register_post_hook(method)`**: Logic to run after a method execution.
 The `modify` method allows you to alter instances of a class at the time they are created or after their creation. This is particularly useful for setting or modifying configuration settings, altering default behaviors, or adding new functionalities to existing classes without changing the original class definitions.
 - **Usage**: Modify a class to include additional configurations or changes
 - **Modify Method Signature**:
  - `modify(modification_method)`: A function that is invoked with an instance of the class as its argument. This function should perform any modifications directly on this instance.
 - **Modification Method Signature**:
  - **Arguments**:
    - `cls`:  This function accepts a single argument, the class instance, which it then modifies.
  - **Modifiable Classes**:
    - `connapp.config`
    - `connapp.node`
    - `connapp.nodes`
    - `connapp.ai`
  - ```python
    def modify_config(cls):
        # Example modification: adding a new attribute or modifying an existing one
        cls.new_attribute = 'New Value'
    class Preload:
        def __init__(self, connapp):
            # Applying modification to the config class instance
            connapp.config.modify(modify_config)
    ```
 #### Implementing Method Hooks
 There are 2 methods that allows you to define custom logic to be executed before (`register_pre_hook`) or after (`register_post_hook`) the main logic of a method. This is particularly useful for logging, auditing, preprocessing inputs, postprocessing outputs or adding functionalities.
  - **Usage**: Register hooks to methods to execute additional logic before or after the main method execution.
 - **Registration Methods Signature**:
  - `register_pre_hook(pre_hook_method)`: A function that is invoked before the main method is executed. This function should do preprocessing of the arguments.
  - `register_post_hook(post_hook_method)`: A function that is invoked after the main method is executed. This function should do postprocessing of the outputs.
 - **Method Signatures for Pre-Hooks**
  - `pre_hook_method(*args, **kwargs)`
  - **Arguments**:
    - `*args`, `**kwargs`: The arguments and keyword arguments that will be passed to the method being hooked. The pre-hook function has the opportunity to inspect and modify these arguments before they are passed to the main method.
  - **Return**:
    - Must return a tuple `(args, kwargs)`, which will be used as the new arguments for the main method. If the original arguments are not modified, the function should return them as received.
 - **Method Signatures for Post-Hooks**:
  - `post_hook_method(*args, **kwargs)`
  - **Arguments**:
    - `*args`, `**kwargs`: The arguments and keyword arguments that were passed to the main method.
      - `kwargs["result"]`: The value returned by the main method. This allows the post-hook to inspect and even alter the result before it is returned to the original caller.
  - **Return**:
    - Can return a modified result, which will replace the original result of the main method, or simply return `kwargs["result"]` to return the original method result.    
  - ```python
    def pre_processing_hook(*args, **kwargs):
        print("Pre-processing logic here")
        # Modify arguments or perform any checks
        return args, kwargs  # Return modified or unmodified args and kwargs
    def post_processing_hook(*args, **kwargs):
        print("Post-processing logic here")
        # Modify the result or perform any final logging or cleanup
        return kwargs["result"]  # Return the modified or unmodified result
    class Preload:
        def __init__(self, connapp):
            # Registering a pre-hook
            connapp.ai.some_method.register_pre_hook(pre_processing_hook)
            # Registering a post-hook
            connapp.node.another_method.register_post_hook(post_processing_hook)
    ```
 ### Executable Block
 - The plugin script can include an executable block:
  - `if __name__ == "__main__":`
  - This block allows the plugin to be run as a standalone script for testing or independent use.
 ### Command Completion Support
 Plugins can provide intelligent tab completion:
 1. **Tree-based Completion (Recommended)**: Define `_connpy_tree(info)` returning a navigation dictionary.
 2. **Legacy Completion**: Define `_connpy_completion(wordsnumber, words, info)`.
-Plugins can provide intelligent **tab completion** by defining autocompletion logic. There are two supported methods, with the tree-based approach being the most modern and recommended.
+---
-#### 1. Tree-based Completion (Recommended)
+## ⚙️ gRPC Service Architecture
 Connpy can operate in a decoupled mode:
 1. **Start the API (Server)**: `conn api -s 50051`
 2. **Configure the Client**:
   ```bash
   conn config --service-mode remote
   conn config --remote-host localhost:50051
   ```
 All inventory management and execution will now happen on the server.
-Define a function called `_connpy_tree` that returns a declarative navigation tree. This method is highly efficient, supports complex state loops, and is very simple to implement for most use cases.
+---
 ## 🐍 Automation Module (API)
 You can use `connpy` as a Python library for your own scripts.
 ### Basic Execution
 ```python
 def _connpy_tree(info=None):
    nodes = info.get("nodes", [])
    return {
        "__exclude_used__": True,  # Filter out words already typed
        "__extra__": nodes,        # Suggest nodes at this level
        "--format": ["json", "yaml", "table"], # Fixed suggestions
        "*": {                     # Wildcard matches any positional word
            "interface1": None,
            "interface2": None,
            "--verbose": None
        }
    }
 ```
 - **Keys**: Literal completions (exact matches).
 - **`*` Key**: A wildcard that matches any positional word typed by the user.
 - **`__extra__`**: A list or a callable `(words) -> list` that adds dynamic suggestions.
 - **`__exclude_used__`**: (Boolean) If True, automatically filters out words already present in the command line.
 #### 2. Legacy Function-based Completion
 For backward compatibility or highly custom logic, you can define `_connpy_completion`.
 ```python
 def _connpy_completion(wordsnumber, words, info=None):
    if wordsnumber == 3:
        return ["--help", "--verbose", "start", "stop"]
    elif wordsnumber == 4 and words[2] == "start":
        return info["nodes"]  # Suggest node names
    return []
 ```
 | Parameter      | Description |
 |----------------|-------------|
 | `wordsnumber`  | Integer indicating the total number of words on the command line. For plugins, this typically starts at 3. |
 | `words`        | A list of tokens (words) already typed. `words[0]` is always the name of the plugin. |
 | `info`         | A dictionary of structured context data (`nodes`, `folders`, `profiles`, `config`). |
 > In this example, if the user types `connpy myplugin start ` and presses Tab, it will suggest node names.
 ### Handling Unknown Arguments
 Plugins can choose to accept and process unknown arguments that are **not explicitly defined** in the parser. To enable this behavior, the plugin must define the following hidden argument in its `Parser` class:
 ```
 self.parser.add_argument(
    "--unknown-args",
    action="store_true",
    default=True,
    help=argparse.SUPPRESS
 )
 ```
 #### Behavior:
 - When this argument is present, Connpy will parse the known arguments and capture any extra (unknown) ones.
 - These unknown arguments will be passed to the plugin as `args.unknown_args` inside the `Entrypoint`.
 - If the user does not pass any unknown arguments, `args.unknown_args` will contain the default value (`True`, unless overridden).
 #### Example:
 If a plugin accepts unknown tcpdump flags like this:
 ```
 connpy myplugin -nn -s0
 ```
 And defines the hidden `--unknown-args` flag as shown above, then:
 - `args.unknown_args` inside `Entrypoint.__init__()` will be: `['-nn', '-s0']`
 > This allows the plugin to receive and process arguments intended for external tools (e.g., `tcpdump`) without argparse raising an error.
 #### Note:
 If a plugin does **not** define `--unknown-args`, any extra arguments passed will cause argparse to fail with an unrecognized arguments error.
 ### Script Verification
 - The `verify_script` method in `plugins.py` is used to check the plugin script's compliance with these standards.
 - Non-compliant scripts will be rejected to ensure consistency and proper functionality within the plugin system.
 - 
 ### Example Script
 For a practical example of how to write a compatible plugin script, please refer to the following example:
 [Example Plugin Script](https://github.com/fluzzi/awspy)
 This script demonstrates the required structure and implementation details according to the plugin system's standards.
 ## gRPC Service Architecture
 Connpy features a completely decoupled gRPC Client/Server architecture. You can run Connpy as a standalone background service and connect to it remotely via the CLI or other clients.
 ### 1. Start the Server
 Start the gRPC service by running:
 ```bash
 connpy api -s 50051
 ```
 The server will handle all configurations, connections, AI sessions, and plugin execution locally on the machine it runs on.
 ### 2. Connect the Client
 Configure your local CLI client to connect to the remote server:
 ```bash
 connpy config --service-mode remote
 connpy config --remote-host localhost:50051
 ```
 Once configured, all commands (`connpy node`, `connpy list`, `connpy ai`, etc.) will execute transparently on the remote server via thin-client proxies. You can revert back to standalone execution at any time by running `connpy config --service-mode local`.
 ### Programmatic Access (gRPC & SOA)
 Developers can build their own applications using the Connpy backend by utilizing the `ServiceProvider`:
 ```python
 from connpy.services.provider import ServiceProvider
 services = ServiceProvider(config, mode="remote", remote_host="localhost:50051")
 nodes = services.nodes.list_nodes()
 ```
 ## Automation module
 The automation module
 ### Standalone module
 ```
 import connpy
-router = connpy.node("uniqueName","ip/host", user="user", password="pass")
+router = connpy.node("uniqueName", "1.1.1.1", user="admin")
-router.run(["term len 0","show run"])
+router.run(["show ip int brief"])
 print(router.output)
 hasip = router.test("show ip int brief","1.1.1.1")
 if hasip:
    print("Router has ip 1.1.1.1")
 else:
    print("router does not have ip 1.1.1.1")
 ```
-### Using manager configuration
+### Parallel Tasks with Variables
-```
+```python
 import connpy
 conf = connpy.configfile()
 device = conf.getitem("server@office")
 server = connpy.node("unique name", **device, config=conf)
 result = server.run(["cd /", "ls -la"])
 print(result)
 ```
 ### Running parallel tasks 
 ```
 import connpy
 conf = connpy.configfile()
 #You can get the nodes from the config from a folder and fitlering in it
 nodes = conf.getitem("@office", ["router1", "router2", "router3"])
 #You can also get each node individually:
 nodes = {}
 nodes["router1"] = conf.getitem("router1@office")
 nodes["router2"] = conf.getitem("router2@office")
 nodes["router10"] = conf.getitem("router10@datacenter")
 #Also, you can create the nodes manually:
 nodes = {}
 nodes["router1"] = {"host": "1.1.1.1", "user": "user", "password": "pass1"}
 nodes["router2"] = {"host": "1.1.1.2", "user": "user", "password": "pass2"}
 nodes["router3"] = {"host": "1.1.1.2", "user": "user", "password": "pass3"}
 #Finally you run some tasks on the nodes
 mynodes = connpy.nodes(nodes, config = conf)
 result = mynodes.test(["show ip int br"], "1.1.1.2")
 for i in result:
    print("---" + i + "---")
    print(result[i])
    print()
 # Or for one specific node
 mynodes.router1.run(["term len 0". "show run"], folder = "/home/user/logs")
 ```
 ### Using variables
 ```
 import connpy
 config = connpy.configfile()
-nodes = config.getitem("@office", ["router1", "router2", "router3"])
+nodes = config.getitem("@office", ["router1", "router2"])
-commands = []
+routers = connpy.nodes(nodes, config=config)
-commands.append("config t")
+
-commands.append("interface lo {id}")
+variables = {
-commands.append("ip add {ip} {mask}")
+    "router1@office": {"id": "1"},
-commands.append("end")
+    "__global__": {"mask": "255.255.255.0"}
-variables = {}
+}
-variables["router1@office"] = {"ip": "10.57.57.1"}
+routers.run(["interface lo{id}", "ip address 10.0.0.{id} {mask}"], variables)
 variables["router2@office"] = {"ip": "10.57.57.2"}
 variables["router3@office"] = {"ip": "10.57.57.3"}
 variables["__global__"] = {"id": "57"}
 variables["__global__"]["mask"] =  "255.255.255.255"
 expected = "!"
 routers = connpy.nodes(nodes, config = config)
 routers.run(commands, variables)
 routers.test("ping {ip}", expected, variables)
 for key in routers.result:
    print(key, ' ---> ', ("pass" if routers.result[key] else "fail"))
 ```
 ### Using AI
 ```
 ### AI Programmatic Use
 ```python
 import connpy
-conf = connpy.configfile()
+myai = connpy.ai(connpy.configfile())
-# Uses models and API keys from config, or override them:
+response = myai.ask("What is the status of the BGP neighbors in the office?")
 myai = connpy.ai(conf, engineer_model="gemini/gemini-2.5-flash", engineer_api_key="your-key")
 result = myai.ask("go to router1 and show me the running configuration")
 print(result["response"])
 # Streaming is enabled by default for CLI, disable for programmatic use:
 result = myai.ask("show interfaces on all routers", stream=False)
 print(result["response"])
 ```
-#### AI Plugin Tool Registration
+---
-Plugins can register custom tools with the AI system using `register_ai_tool()` in their `Preload` class:
+*For detailed developer notes and plugin hooks documentation, see the [Documentation](https://fluzzi.github.io/connpy/).*
 ```
 def _register_my_tools(ai_instance):
    tool_def = {
        "type": "function",
        "function": {
            "name": "my_custom_tool",
            "description": "Does something useful.",
            "parameters": {
                "type": "object",
                "properties": {"query": {"type": "string"}},
                "required": ["query"]
            }
        }
    }
    ai_instance.register_ai_tool(
        tool_definition=tool_def,
        handler=my_handler_function,
        target="engineer",  # or "architect" or "both"
        engineer_prompt="- My tool: does X.",
        architect_prompt="  * My tool (my_custom_tool)."
    )
 class Preload:
    def __init__(self, connapp):
        connapp.ai.modify(_register_my_tools)
 ```
 ## Developer Notes (SOA Architecture)
 As of version 2.0, Connpy has migrated to a **Service-Oriented Architecture (SOA)**:
 - **`connpy/cli/`**: Contains all CLI handlers. These are responsible for argument parsing, user interaction (via `inquirer`), and visual output (via `printer`).
 - **`connpy/services/`**: Contains pure logic services (Node, Profile, Execution, etc.).
 - **Zero-Print Policy**: Services must never use `print()`. All output must be returned as data structures or generators to the caller (CLI handlers).
 - **ServiceProvider**: Access services via `connapp.services`. This allows transparent switching between local and remote (gRPC) backends without modifying CLI logic.
 '''
 from .core import node,nodes
 from .configfile import configfile
@@ -1 +1 @@
-__version__ = "6.0.0b5"
+__version__ = "6.0.0b11"
@@ -3,8 +3,11 @@ import sys
 import json
 import re
 import datetime
 import threading
 import asyncio
 from textwrap import dedent
 from .core import nodes
 from .mcp_client import MCPClientManager
 _litellm_initialized = False
@@ -37,6 +40,61 @@ from rich.rule import Rule
 console = printer.console
 _ai_loop = None
 _ai_thread = None
 _ai_lock = threading.Lock()
 def _get_ai_loop():
    global _ai_loop, _ai_thread
    with _ai_lock:
        if _ai_loop is None:
            _ai_loop = asyncio.new_event_loop()
            _ai_thread = threading.Thread(target=_ai_loop.run_forever, name="ConnpyAILoop", daemon=True)
            _ai_thread.start()
        return _ai_loop
 def run_ai_async(coro):
    """Run a coroutine in the dedicated AI background loop."""
    loop = _get_ai_loop()
    return asyncio.run_coroutine_threadsafe(coro, loop)
 def cleanup():
    """Safely close any global litellm sessions in the dedicated AI loop."""
    global _ai_loop
    if _ai_loop:
        try:
            future = asyncio.run_coroutine_threadsafe(_async_cleanup(), _ai_loop)
            future.result(timeout=5)
        except:
            pass
 async def _async_cleanup():
    """Internal async cleanup for litellm sessions."""
    try:
        import litellm
        # 1. Close synchronous session
        if hasattr(litellm, "client_session") and litellm.client_session:
            try:
                if hasattr(litellm.client_session, "close"):
                    res = litellm.client_session.close()
                    if asyncio.iscoroutine(res): await res
            except: pass
            litellm.client_session = None
        # 2. Close asynchronous session
        if hasattr(litellm, "aclient_session") and litellm.aclient_session:
            try:
                session = litellm.aclient_session
                litellm.aclient_session = None
                if hasattr(session, "close"):
                    await session.close()
            except: pass
    except ImportError:
        pass
@ClassHook
 class ai:
    """Hybrid Multi-Agent System: Selective Escalation with Role Persistence."""
@@ -60,7 +118,7 @@ class ai:
        aiconfig = self.config.config.get("ai", {})
        # Modelos (Prioridad: Argumento -> Config -> Default)
-        self.engineer_model = engineer_model or aiconfig.get("engineer_model") or "gemini/gemini-3.1-flash-lite-preview"
+        self.engineer_model = engineer_model or aiconfig.get("engineer_model") or "gemini/gemini-3.1-flash-lite"
        self.architect_model = architect_model or aiconfig.get("architect_model") or "anthropic/claude-sonnet-4-6"
        # API Keys (Prioridad: Argumento -> Config)
@@ -87,6 +145,9 @@ class ai:
        self.engineer_prompt_extensions = []  # Extra text for engineer prompt
        self.architect_prompt_extensions = [] # Extra text for architect prompt
        # MCP Manager
        self.mcp_manager = MCPClientManager(self.config)
        # Long-term memory
        self.memory_path = os.path.join(self.config.defaultdir, "ai_memory.md")
        self.long_term_memory = ""
@@ -144,6 +205,7 @@ class ai:
            - COMPLETE MISSIONS: Execute ALL steps of a mission before reporting back.
            - DIAGRAM: Use ASCII art or Unicode box-drawing characters directly in your responses to visualize topologies or paths when helpful.
            - EVIDENCE: Include 'Key Snippets' from tool outputs. Be token-efficient.
            - LANGUAGE: You MUST respond in the same language used by the user in their question or instruction.
            - NO WANDERING: Do not speculate. If stuck, report attempts.
            - SAFETY: When you use 'run_commands' with configuration commands, the system automatically prompts the user for confirmation. Just execute - don't ask permission first.
 {architect_instructions}
@@ -161,6 +223,7 @@ class ai:
            - ENGINEER CAPABILITIES: Your Engineer can:
                * Filter nodes (list_nodes), Run CLI commands (run_commands), Get metadata (get_node_info).
            - ANALYSIS: Review technical findings to identify patterns or design failures.
            - LANGUAGE: You MUST respond in the same language used by the user in their question or instruction.
            - MEMORY: Update long-term facts ONLY when the user explicitly requests it.
            CRITICAL - EFFICIENT DELEGATION:
@@ -236,7 +299,6 @@ class ai:
        - response: reconstructed ModelResponse (same as non-streaming)
        - streamed: True if text was rendered to console during streaming
        """
        from rich.live import Live
        stream_resp = completion(model=model, messages=messages, tools=tools, api_key=api_key, stream=True, **kwargs)
@@ -244,7 +306,7 @@ class ai:
        full_content = ""
        is_streaming_text = False
        has_tool_calls = False
-        live_display = None
+        header_printed = False
        # Determine styling based on current brain
        role_label = "Network Architect" if "architect" in label.lower() else "Network Engineer"
@@ -273,7 +335,6 @@ class ai:
                    if not chunk_callback:
                        if not is_streaming_text:
                            # Stop spinner definitively
                            if status:
                                try:
                                    status.stop()
@@ -282,35 +343,28 @@ class ai:
                            # Create a stable, direct Console to bypass _ConsoleProxy recreation bugs
                            from rich.console import Console as RichConsole
-                            from .printer import connpy_theme, get_original_stdout
+                            from rich.rule import Rule
                            from .printer import connpy_theme, get_original_stdout, IncrementalMarkdownParser
                            stable_console = RichConsole(theme=connpy_theme, file=get_original_stdout())
-                            live_display = Live(
+                            stable_console.print(Rule(f"[bold {border}]{title}[/bold {border}]", style=border))
-                                Panel(Markdown(full_content), title=title, border_style=border, expand=False),
+                            header_printed = True
-                                console=stable_console,
+                            md_parser = IncrementalMarkdownParser(console=stable_console)
                                refresh_per_second=8,
                                transient=False
                            )
                            live_display.start()
                            is_streaming_text = True
-                        else:
+                        
-                            live_display.update(
+                        md_parser.feed(delta.content)
                                Panel(Markdown(full_content), title=title, border_style=border, expand=False)
                            )
        except Exception as e:
            if not chunks:
                raise
        finally:
-            if live_display:
+            if header_printed:
                # Render final state with complete content
                try:
-                    live_display.update(
+                    md_parser.flush()
-                        Panel(Markdown(full_content), title=title, border_style=border, expand=False)
+                    from rich.console import Console as RichConsole
-                    )
+                    from rich.rule import Rule
-                except Exception:
+                    from .printer import connpy_theme, get_original_stdout
-                    pass
+                    stable_console = RichConsole(theme=connpy_theme, file=get_original_stdout())
-                try:
+                    stable_console.print(Rule(style=border))
                    live_display.stop()
                except Exception:
                    pass
@@ -620,7 +674,7 @@ class ai:
                    self.console.print("[pass]✓ Trust Mode Enabled. All future commands in this session will execute without confirmation.[/pass]")
                elif user_resp_lower in ['y', 'yes']:
                    self.console.print("[pass]✓ Executing...[/pass]")
-                elif user_resp_lower in ['n', 'no', '']:
+                elif user_resp_lower in ['n', 'no', '', 'cancel']:
                    self.console.print("[fail]✗ Execution rejected by user.[/fail]")
                    return "Error: User rejected execution."
                else:
@@ -716,6 +770,10 @@ class ai:
                            cmd_str = cmds[0] if cmds else ""
                            status.update(f"[ai_status]Engineer: [CMD] {cmd_str}")
                        elif fn == "get_node_info": status.update(f"[ai_status]Engineer: [INSPECT] {args.get('node_name','')}")
                        elif fn.startswith("mcp_"):
                            server = fn.split("__")[0].replace("mcp_", "")
                            tool = fn.split("__")[1] if "__" in fn else fn
                            status.update(f"[ai_status]Engineer: [MCP:{server}] {tool}")
                        elif fn in self.tool_status_formatters: status.update(self.tool_status_formatters[fn](args))
                    if debug:
@@ -724,6 +782,8 @@ class ai:
                    if fn == "list_nodes": obs = self.list_nodes_tool(**args)
                    elif fn == "run_commands": obs = self.run_commands_tool(**args, status=status)
                    elif fn == "get_node_info": obs = self.get_node_info_tool(**args)
                    elif fn.startswith("mcp_"):
                        obs = run_ai_async(self.mcp_manager.call_tool(fn, args)).result(timeout=60)
                    elif fn in self.external_tool_handlers: obs = self.external_tool_handlers[fn](self, **args)
                    else: obs = f"Error: Unknown tool '{fn}'."
@@ -744,14 +804,22 @@ class ai:
        except Exception as e:
            return f"Engineer failed: {str(e)}", usage
-    def _get_engineer_tools(self):
+    def _get_engineer_tools(self, os_filter: str = None):
        """Define tools available to the Engineer."""
        base_tools = [
-            {"type": "function", "function": {"name": "list_nodes", "description": "Lists available nodes in the inventory.", "parameters": {"type": "object", "properties": {"filter_pattern": {"type": "string", "description": "Regex to filter nodes (e.g. '.*', 'border.*')."}}}}},
+            {"type": "function", "function": {"name": "list_nodes", "description": "[Universal Platform] Lists available nodes in the inventory.", "parameters": {"type": "object", "properties": {"filter_pattern": {"type": "string", "description": "Regex to filter nodes (e.g. '.*', 'border.*')."}}}}},
-            {"type": "function", "function": {"name": "run_commands", "description": "Runs one or more commands on matched nodes. MANDATORY: You MUST call 'list_nodes' first to verify the target list.", "parameters": {"type": "object", "properties": {"nodes_filter": {"type": "string", "description": "Exact node name or verified filter pattern."}, "commands": {"type": "array", "items": {"type": "string"}, "description": "List of commands (e.g. ['show ip route', 'show int desc'])."}}, "required": ["nodes_filter", "commands"]}}},
+            {"type": "function", "function": {"name": "run_commands", "description": "[Universal Platform] Runs one or more commands on matched nodes. MANDATORY: You MUST call 'list_nodes' first to verify the target list.", "parameters": {"type": "object", "properties": {"nodes_filter": {"type": "string", "description": "Exact node name or verified filter pattern."}, "commands": {"type": "array", "items": {"type": "string"}, "description": "List of commands (e.g. ['show ip route', 'show int desc'])."}}, "required": ["nodes_filter", "commands"]}}},
-            {"type": "function", "function": {"name": "get_node_info", "description": "Gets full metadata for a specific node.", "parameters": {"type": "object", "properties": {"node_name": {"type": "string"}}, "required": ["node_name"]}}}
+            {"type": "function", "function": {"name": "get_node_info", "description": "[Universal Platform] Gets full metadata for a specific node.", "parameters": {"type": "object", "properties": {"node_name": {"type": "string"}}, "required": ["node_name"]}}}
        ]
        # Add dynamic tools from MCP
        try:
            mcp_tools = run_ai_async(self.mcp_manager.get_tools_for_llm(os_filter=os_filter)).result(timeout=10)
            base_tools.extend(mcp_tools)
        except Exception as e:
            # Silently fail for LLM tools
            pass
        if self.architect_key:
            base_tools.extend([
                {"type": "function", "function": {"name": "consult_architect", "description": "Ask the Strategic Reasoning Engine for advice on complex design, architecture, or troubleshooting decisions. You remain in control and will present the response to the user. Use this for: configuration planning, design validation, complex troubleshooting.", "parameters": {"type": "object", "properties": {"question": {"type": "string", "description": "Strategic question or decision needed."}, "technical_summary": {"type": "string", "description": "Technical findings and context gathered so far."}}, "required": ["question", "technical_summary"]}}},
@@ -1145,6 +1213,8 @@ class ai:
                    elif fn == "run_commands": obs = self.run_commands_tool(**args, status=status)
                    elif fn == "get_node_info": obs = self.get_node_info_tool(**args)
                    elif fn == "manage_memory_tool": obs = self.manage_memory_tool(**args)
                    elif fn.startswith("mcp_"):
                        obs = run_ai_async(self.mcp_manager.call_tool(fn, args)).result(timeout=60)
                    elif fn in self.external_tool_handlers: obs = self.external_tool_handlers[fn](self, **args)
                    else: obs = f"Error: {fn} unknown."
@@ -1211,5 +1281,279 @@ class ai:
            "streamed": streamed_response
        }
    @MethodHook
    async def aask_copilot(self, terminal_buffer, user_question, node_info=None, chunk_callback=None):
        import json
        import re
        from litellm import acompletion
        import asyncio
        import warnings
        import aiohttp
        # Suppress unawaited coroutine warnings from LiteLLM's internal streaming logic during sudden cancellation
        warnings.filterwarnings("ignore", message="coroutine '.*async_streaming.*' was never awaited", category=RuntimeWarning)
        node_info = node_info or {}
        os_info = node_info.get("os", "unknown")
        node_name = node_info.get("name", "unknown")
        persona = node_info.get("persona", "engineer")
        memories = node_info.get("memories", [])
        vendor_reference = ""
        if os_info and os_info != "unknown":
            try:
                os_filename = os_info.lower().replace(" ", "_")
                ref_path = os.path.join(self.config.defaultdir, "ai_references", f"{os_filename}.md")
                if os.path.exists(ref_path):
                    with open(ref_path, "r") as f:
                        vendor_reference = f.read().strip()
            except Exception:
                pass
        if persona == "architect":
            system_prompt = f"""Role: NETWORK ARCHITECT. You act as a senior strategic advisor during a live SSH session.
 Rules:
 1. MANDATORY: You MUST respond in the same language used by the user in their question.
 2. Answer the user's question directly and EXCLUSIVELY based on the Terminal Context. 
 3. NO HALLUCINATIONS. The Terminal Context is a live buffer. If it contains only a shell prompt (like 'iol#' or 'admin@vrouter>') and no command output, it means YOU DON'T HAVE DATA. In this case, YOU MUST NOT invent any information.
 4. Focus on the "why" and "how". Analyze topologies, design patterns, and validate configurations.
 5. Do NOT provide commands to execute unless specifically requested. Instead, explain the consequences and best practices.
 6. Keep your guide concise and authoritative.
 7. You MUST output your response in the following strict format:
 <guide>
 Your brief tactical guide in markdown.
 </guide>
 <commands>
 </commands>
 <risk>
 low
 </risk>
 8. Risk level is usually "low" for read-only/no commands.
 Terminal Context:
 {terminal_buffer}
 Device OS: {os_info}
 Node: {node_name}"""
        else:
            system_prompt = f"""Role: TERMINAL COPILOT. You assist a network engineer during a live SSH session.
 Rules:
 1. MANDATORY: You MUST respond in the same language used by the user in their question.
 2. EXTREMELY IMPORTANT: Answer EXCLUSIVELY based on the provided Terminal Context. 
 3. NO HALLUCINATIONS. The Terminal Context is a live buffer. If it contains only a shell prompt (like 'iol#' or 'admin@vrouter>') and no command output, it means YOU DON'T HAVE DATA. In this case, YOU MUST NOT invent any information. Instead, explicitly state that you don't see the data and offer the correct CLI commands to retrieve it.
 4. If the user asks you to analyze, parse, or extract data from the Terminal Context, DO IT directly in the <guide> section (you can use markdown tables or lists). Do NOT just give them a command to do it themselves.
 5. If the user wants to execute an action, provide the required CLI commands inside a <commands> block, one command per line. If no commands are needed, leave it empty or omit the block.
 6. ULTRA-CONCISE. Keep your guide to the point.
 7. You MUST output your response in the following strict format:
 <guide>
 Your brief tactical guide in markdown. 3-4 sentences max.
 </guide>
 <commands>
 command 1
 command 2
 </commands>
 <risk>
 low, high, or destructive
 </risk>
 8. Risk level: "low" for read-only/no commands, "high" for config changes, "destructive" for potentially dangerous ops.
 Terminal Context:
 {terminal_buffer}
 Device OS: {os_info}
 Node: {node_name}"""
        if vendor_reference:
            system_prompt += f"\n\nVendor Command Reference:\n{vendor_reference}"
        if memories:
            system_prompt += "\n\nSession Memory (Important Facts):\n"
            for m in memories:
                system_prompt += f"- {m}\n"
        # Fetch MCP tools for the current OS
        mcp_tools = []
        try:
            mcp_tools = await self.mcp_manager.get_tools_for_llm(os_filter=os_info)
        except Exception:
            pass
        if mcp_tools:
            system_prompt += f"\n\nAvailable MCP Tools: {', '.join([t['function']['name'] for t in mcp_tools])}"
            system_prompt += "\nUse these tools to validate syntax or find exact commands if needed before providing the final guide."
        messages = [
            {"role": "system", "content": system_prompt},
            {"role": "user", "content": user_question}
        ]
        iteration = 0
        max_iterations = 5 # Allow up to 5 iterations for tool usage
        # Use models based on persona
        current_model = self.architect_model if persona == "architect" else self.engineer_model
        current_key = self.architect_key if persona == "architect" else self.engineer_key
        try:
            while iteration < max_iterations:
                iteration += 1
                response = await acompletion(
                    model=current_model,
                    messages=messages,
                    tools=mcp_tools if mcp_tools else None,
                    api_key=current_key,
                    stream=True
                )
                full_content = ""
                streamed_guide = ""
                tool_calls = []
                async for chunk in response:
                    delta = chunk.choices[0].delta
                    # Accumulate tool calls
                    if hasattr(delta, 'tool_calls') and delta.tool_calls:
                        for tc in delta.tool_calls:
                            idx = tc.index
                            if idx >= len(tool_calls):
                                tool_calls.append({"id": tc.id, "type": "function", "function": {"name": tc.function.name or "", "arguments": tc.function.arguments or ""}})
                            else:
                                if tc.id: tool_calls[idx]["id"] = tc.id
                                if tc.function.name: tool_calls[idx]["function"]["name"] = tc.function.name
                                if tc.function.arguments: tool_calls[idx]["function"]["arguments"] += tc.function.arguments
                    if hasattr(delta, 'content') and delta.content:
                        full_content += delta.content
                        if chunk_callback and not tool_calls: # Only stream if not using tools
                            start_idx = full_content.find("<guide>")
                            if start_idx != -1:
                                after_start = full_content[start_idx + 7:]
                                end_idx = after_start.find("</guide>")
                                if end_idx != -1:
                                    current_guide = after_start[:end_idx]
                                else:
                                    current_guide = after_start
                                    if current_guide.endswith("<"): current_guide = current_guide[:-1]
                                    elif current_guide.endswith("</"): current_guide = current_guide[:-2]
                                    elif current_guide.endswith("</g"): current_guide = current_guide[:-3]
                                    elif current_guide.endswith("</gu"): current_guide = current_guide[:-4]
                                    elif current_guide.endswith("</gui"): current_guide = current_guide[:-5]
                                    elif current_guide.endswith("</guid"): current_guide = current_guide[:-6]
                                    elif current_guide.endswith("</guide"): current_guide = current_guide[:-7]
                                new_text = current_guide[len(streamed_guide):]
                                if new_text:
                                    chunk_callback(new_text)
                                    streamed_guide += new_text
                if not tool_calls:
                    break
                # Execute tool calls
                messages.append({"role": "assistant", "content": full_content or None, "tool_calls": tool_calls})
                for tc in tool_calls:
                    fn = tc["function"]["name"]
                    args = json.loads(tc["function"]["arguments"])
                    if "mcp_" in fn:
                        try:
                            obs = await asyncio.wait_for(self.mcp_manager.call_tool(fn, args), timeout=30.0)
                        except Exception as e:
                            obs = f"Error calling MCP tool: {e}"
                    else:
                        obs = f"Error: Tool {fn} not allowed in Copilot."
                    messages.append({"tool_call_id": tc["id"], "role": "tool", "name": fn, "content": self._truncate(str(obs))})
            # If we hit the limit and it was still using tools, force a final answer
            if tool_calls and iteration >= max_iterations:
                messages.append({"role": "user", "content": "Tool limit reached. Provide your final tactical guide now based on the findings."})
                response = await acompletion(
                    model=self.engineer_model,
                    messages=messages,
                    tools=None,
                    api_key=self.engineer_key,
                    stream=True
                )
                full_content = ""
                streamed_guide = ""
                async for chunk in response:
                    delta = chunk.choices[0].delta
                    if hasattr(delta, 'content') and delta.content:
                        full_content += delta.content
                        if chunk_callback:
                            start_idx = full_content.find("<guide>")
                            if start_idx != -1:
                                after_start = full_content[start_idx + 7:]
                                end_idx = after_start.find("</guide>")
                                if end_idx != -1:
                                    current_guide = after_start[:end_idx]
                                else:
                                    current_guide = after_start
                                    if current_guide.endswith("<"): current_guide = current_guide[:-1]
                                    elif current_guide.endswith("</"): current_guide = current_guide[:-2]
                                    elif current_guide.endswith("</g"): current_guide = current_guide[:-3]
                                    elif current_guide.endswith("</gu"): current_guide = current_guide[:-4]
                                    elif current_guide.endswith("</gui"): current_guide = current_guide[:-5]
                                    elif current_guide.endswith("</guid"): current_guide = current_guide[:-6]
                                    elif current_guide.endswith("</guide"): current_guide = current_guide[:-7]
                                new_text = current_guide[len(streamed_guide):]
                                if new_text:
                                    chunk_callback(new_text)
                                    streamed_guide += new_text
            guide = ""
            commands = []
            risk_level = "low"
            guide_match = re.search(r"<guide>(.*?)</guide>", full_content, re.DOTALL)
            if guide_match:
                guide = guide_match.group(1).strip()
            cmd_match = re.search(r"<commands>(.*?)</commands>", full_content, re.DOTALL)
            if cmd_match:
                cmds_raw = cmd_match.group(1).strip()
                if cmds_raw:
                    commands = [c.strip() for c in cmds_raw.split('\n') if c.strip()]
            risk_match = re.search(r"<risk>(.*?)</risk>", full_content, re.DOTALL)
            if risk_match:
                risk_level = risk_match.group(1).strip().lower()
            if not guide and full_content and not ("<guide>" in full_content):
                guide = full_content.strip()
            return {
                "commands": commands,
                "guide": guide,
                "risk_level": risk_level,
                "error": None
            }
        except asyncio.CancelledError:
            # Client cancelled the request via gRPC or local interrupt
            if 'response' in locals():
                try:
                    if hasattr(response, 'aclose'):
                        # Fire and forget the close to avoid blocking the cancel
                        asyncio.create_task(response.aclose())
                    elif hasattr(response, 'close'):
                        response.close()
                except Exception:
                    pass
            return None
        except Exception as e:
            return {
                "commands": [],
                "guide": "",
                "risk_level": "low",
                "error": str(e)
            }
    @MethodHook
    def confirm(self, user_input): return True
@@ -54,6 +54,8 @@ def debug_api(port=8048, config=None):
    printer.info(f"gRPC Server running in debug mode on port {port}...")
    _wait_for_termination()
    server.stop(0)
    from .ai import cleanup
    cleanup()
 def start_server(port=8048, config=None):
    try:
@@ -67,6 +69,9 @@ def start_server(port=8048, config=None):
        conf = config or configfile()
        server = serve(conf, port=port, debug=False)
        _wait_for_termination()
        server.stop(0)
        from .ai import cleanup
        cleanup()
    except Exception as e:
        printer.error(f"Background API failed to start: {e}")
        os._exit(1)
@@ -32,6 +32,9 @@ class AIHandler:
                printer.error(str(e))
            return
        if args.mcp is not None:
            return self.configure_mcp(args)
        # Determinar session_id para retomar
        session_id = None
        if args.resume:
@@ -110,7 +113,7 @@ class AIHandler:
            try:
                user_query = Prompt.ask("[user_prompt]User[/user_prompt]")
                if not user_query.strip(): continue
-                if user_query.lower() in ['exit', 'quit', 'bye']: break
+                if user_query.lower() in ['exit', 'quit', 'bye', 'cancel']: break
                with console.status("[ai_status]Agent is thinking...") as status:
                    result = self.app.myai.ask(user_query, chat_history=history, status=status, debug=args.debug, trust=args.trust, **self.ai_overrides)
@@ -134,3 +137,115 @@ class AIHandler:
            except (KeyboardInterrupt, EOFError):
                console.print("\n[dim]Session closed.[/dim]")
                break
    def configure_mcp(self, args):
        """Handle MCP server configuration via CLI tokens or interactive wizard."""
        mcp_args = args.mcp
        # 1. Non-interactive CLI Mode (if arguments are provided)
        if mcp_args:
            action = mcp_args[0].lower()
            if action == "list":
                settings = self.app.services.config_svc.get_settings()
                mcp_servers = settings.get("ai", {}).get("mcp_servers", {})
                if not mcp_servers:
                    printer.info("No MCP servers configured.")
                else:
                    columns = ["Name", "URL", "Enabled", "Auto-load OS"]
                    rows = []
                    for name, cfg in mcp_servers.items():
                        rows.append([
                            name, 
                            cfg.get("url", ""), 
                            "[green]Yes[/green]" if cfg.get("enabled", True) else "[red]No[/red]",
                            cfg.get("auto_load_on_os", "Any")
                        ])
                    printer.table("Configured MCP Servers", columns, rows)
                return
            elif action == "add":
                if len(mcp_args) < 3:
                    printer.error("Usage: connpy ai --mcp add <name> <url> [os_filter]")
                    return
                name, url = mcp_args[1], mcp_args[2]
                os_filter = mcp_args[3] if len(mcp_args) > 3 else None
                try:
                    self.app.services.ai.configure_mcp(name, url=url, auto_load_on_os=os_filter)
                    printer.success(f"MCP server '{name}' added/updated.")
                except Exception as e:
                    printer.error(str(e))
                return
            elif action == "remove":
                if len(mcp_args) < 2:
                    printer.error("Usage: connpy ai --mcp remove <name>")
                    return
                name = mcp_args[1]
                try:
                    self.app.services.ai.configure_mcp(name, remove=True)
                    printer.success(f"MCP server '{name}' removed.")
                except Exception as e:
                    printer.error(str(e))
                return
            elif action in ["enable", "disable"]:
                if len(mcp_args) < 2:
                    printer.error(f"Usage: connpy ai --mcp {action} <name>")
                    return
                name = mcp_args[1]
                enabled = (action == "enable")
                try:
                    self.app.services.ai.configure_mcp(name, enabled=enabled)
                    printer.success(f"MCP server '{name}' {'enabled' if enabled else 'disabled'}.")
                except Exception as e:
                    printer.error(str(e))
                return
            else:
                printer.error(f"Unknown MCP action: {action}")
                printer.info("Available actions: list, add, remove, enable, disable")
                return
        # 2. Interactive Wizard Mode (if no arguments provided)
        # Import forms dynamically to avoid circular dependencies if any
        if not hasattr(self.app, "cli_forms"):
            from .forms import Forms
            self.app.cli_forms = Forms(self.app)
        settings = self.app.services.config_svc.get_settings()
        mcp_servers = settings.get("ai", {}).get("mcp_servers", {})
        result = self.app.cli_forms.mcp_wizard(mcp_servers)
        if not result:
            return
        action = result["action"]
        try:
            if action == "list":
                # Recursive call to the non-interactive list logic
                args.mcp = ["list"]
                return self.configure_mcp(args)
            elif action == "add":
                self.app.services.ai.configure_mcp(
                    result["name"], 
                    url=result["url"], 
                    enabled=result["enabled"],
                    auto_load_on_os=result["os"]
                )
                printer.success(f"MCP server '{result['name']}' saved.")
            elif action == "update": # Used for toggle
                self.app.services.ai.configure_mcp(
                    result["name"], 
                    enabled=result["enabled"]
                )
                printer.success(f"MCP server '{result['name']}' updated.")
            elif action == "remove":
                self.app.services.ai.configure_mcp(result["name"], remove=True)
                printer.success(f"MCP server '{result['name']}' removed.")
        except Exception as e:
            printer.error(str(e))
@@ -197,3 +197,84 @@ class Forms:
            answer["tags"] = ast.literal_eval(answer["tags"])
        return answer
    def mcp_wizard(self, mcp_servers):
        """Interactive wizard to manage MCP servers."""
        from .helpers import theme
        while True:
            options = [
                ("List Configured Servers", "list"),
                ("Add/Update Server", "add"),
                ("Enable/Disable Server", "toggle"),
                ("Remove Server", "remove"),
                ("Back", "exit")
            ]
            questions = [
                inquirer.List("action", message="MCP Configuration", choices=options)
            ]
            answers = inquirer.prompt(questions, theme=theme)
            if not answers or answers["action"] == "exit":
                return None
            action = answers["action"]
            if action == "list":
                if not mcp_servers:
                    print("\nNo MCP servers configured.\n")
                else:
                    return {"action": "list"}
            elif action == "add":
                questions = [
                    inquirer.Text("name", message="Server Name (identifier)"),
                    inquirer.Text("url", message="SSE URL (e.g., http://localhost:8000/sse)"),
                    inquirer.Confirm("enabled", message="Enabled?", default=True),
                    inquirer.Text("auto_load_os", message="Auto-load on specific OS (blank for any)")
                ]
                answers = inquirer.prompt(questions, theme=theme)
                if answers:
                    return {
                        "action": "add",
                        "name": answers["name"],
                        "url": answers["url"],
                        "enabled": answers["enabled"],
                        "os": answers["auto_load_os"]
                    }
            elif action == "toggle":
                if not mcp_servers:
                    print("\nNo servers to toggle.\n")
                    continue
                choices = []
                for name, cfg in mcp_servers.items():
                    status = "[Enabled]" if cfg.get("enabled", True) else "[Disabled]"
                    choices.append((f"{name} {status}", name))
                questions = [
                    inquirer.List("name", message="Select server to toggle", choices=choices + [("Cancel", None)])
                ]
                answers = inquirer.prompt(questions, theme=theme)
                if answers and answers["name"]:
                    current = mcp_servers[answers["name"]].get("enabled", True)
                    return {
                        "action": "update",
                        "name": answers["name"],
                        "enabled": not current
                    }
            elif action == "remove":
                if not mcp_servers:
                    print("\nNo servers to remove.\n")
                    continue
                questions = [
                    inquirer.List("name", message="Select server to remove", choices=list(mcp_servers.keys()) + ["Cancel"])
                ]
                answers = inquirer.prompt(questions, theme=theme)
                if answers and answers["name"] != "Cancel":
                    return {"action": "remove", "name": answers["name"]}
        return None
@@ -121,7 +121,7 @@ class RunHandler:
                    commands=commands,
                    variables=variables,
                    parallel=options.get("parallel", 10),
-                    timeout=options.get("timeout", 10),
+                    timeout=options.get("timeout", 20),
                    folder=folder,
                    prompt=prompt,
                    on_node_complete=_on_run_complete
@@ -155,7 +155,7 @@ class RunHandler:
                    expected=expected,
                    variables=variables,
                    parallel=options.get("parallel", 10),
-                    timeout=options.get("timeout", 10),
+                    timeout=options.get("timeout", 20),
                    folder=folder,
                    prompt=prompt,
                    on_node_complete=_on_test_complete
@@ -0,0 +1,497 @@
 import os
 import re
 import sys
 import time
 import asyncio
 import fcntl
 import termios
 import tty
 from typing import Any, Dict, List, Optional, Callable
 from textwrap import dedent
 from rich.console import Console
 from rich.panel import Panel
 from rich.markdown import Markdown
 from prompt_toolkit import PromptSession
 from prompt_toolkit.key_binding import KeyBindings
 from prompt_toolkit.formatted_text import HTML
 from prompt_toolkit.history import InMemoryHistory
 from ..printer import connpy_theme
 from connpy.utils import log_cleaner
 from ..services.ai_service import AIService
 class CopilotInterface:
    def __init__(self, config, history=None, pt_input=None, pt_output=None, rich_file=None, session_state=None):
        self.config = config
        self.history = history or InMemoryHistory()
        self.pt_input = pt_input
        self.pt_output = pt_output
        self.ai_service = AIService(config)
        self.session_state = session_state if session_state is not None else {
            'persona': 'engineer',
            'trust_mode': False,
            'memories': [],
            'os': None,
            'prompt': None
        }
        if rich_file:
            self.console = Console(theme=connpy_theme, force_terminal=True, file=rich_file)
        else:
            self.console = Console(theme=connpy_theme)
        self.mode_range, self.mode_single, self.mode_lines = 0, 1, 2 
    def _get_theme_color(self, style_name: str, fallback: str = "white") -> str:
        """Extract Hex or ANSI color name from the active rich theme."""
        try:
            style = connpy_theme.styles.get(style_name)
            if style and style.color:
                # If it's a standard color like 'green', Rich might return its hex triplet
                if style.color.is_default: return fallback
                return style.color.triplet.hex if style.color.triplet else style.color.name
        except: pass
        return fallback
    async def run_session(self, 
                          raw_bytes: bytes, 
                          node_info: dict,
                          on_ai_call: Callable,
                          cmd_byte_positions: List[tuple] = None, 
                          blocks: List[tuple] = None):
        """
        Runs the interactive Copilot session.
        on_ai_call: async function(active_buffer, question) -> result_dict
        """
        from rich.rule import Rule
        try:
            # Prepare UI state
            buffer = log_cleaner(raw_bytes.decode(errors='replace'))
            # Use pre-calculated blocks if provided (remote mode), otherwise calculate locally (local mode)
            if blocks is None:
                last_line = buffer.split('\n')[-1].strip() if buffer.strip() else "(prompt)"
                blocks = self.ai_service.build_context_blocks(raw_bytes, cmd_byte_positions, node_info, last_line=last_line)
            state = {
                'context_cmd': 1,
                'total_cmds': len(blocks),
                'total_lines': len(buffer.split('\n')),
                'context_lines': min(50, len(buffer.split('\n'))),
                'context_mode': self.mode_range,
                'cancelled': False,
                'toolbar_msg': '',
                'msg_expiry': 0
            }
            # 1. Visual Separation
            self.console.print("") # Salto de línea real
            self.console.print(Rule(title="[bold cyan] AI TERMINAL COPILOT [/bold cyan]", style="cyan"))
            self.console.print(Panel(
                "[dim]Type your question. Enter to send, Escape/Ctrl+C to cancel. Type / for commands.\n"
                "Tab to change context mode. Ctrl+\u2191/\u2193 to adjust context. \u2191\u2193 for question history.[/dim]",
                border_style="cyan"
            ))
            self.console.print("\n") # Pequeño espacio antes del prompt del copilot
            bindings = KeyBindings()
            @bindings.add('c-up')
            def _(event):
                if state['context_mode'] == self.mode_lines:
                    state['context_lines'] = min(state['context_lines'] + 50, state['total_lines'])
                else:
                    state['context_cmd'] = min(state['context_cmd'] + 1, state['total_cmds'])
                event.app.invalidate()
            @bindings.add('c-down')
            def _(event):
                if state['context_mode'] == self.mode_lines:
                    state['context_lines'] = max(state['context_lines'] - 50, min(50, state['total_lines']))
                else:
                    state['context_cmd'] = max(state['context_cmd'] - 1, 1)
                event.app.invalidate()
            @bindings.add('tab')
            def _(event):
                buf = event.current_buffer
                # If typing a slash command (no spaces yet), use tab to autocomplete inline
                if buf.text.startswith('/') and ' ' not in buf.text:
                    buf.complete_next()
                else:
                    state['context_mode'] = (state['context_mode'] + 1) % 3
                    event.app.invalidate()
            @bindings.add('escape', eager=True)
            @bindings.add('c-c')
            def _(event):
                state['cancelled'] = True
                event.app.exit(result='')
            def get_active_buffer():
                if state['context_mode'] == self.mode_lines:
                    return '\n'.join(buffer.split('\n')[-state['context_lines']:])
                idx = max(0, state['total_cmds'] - state['context_cmd'])
                start, end, preview = blocks[idx]
                if state['context_mode'] == self.mode_single:
                    active_raw = raw_bytes[start:end]
                else:
                    active_raw = raw_bytes[start:]
                return preview + "\n" + log_cleaner(active_raw.decode(errors='replace'))
            def get_prompt_text():
                import html
                # Always use user_prompt color for the Ask prompt
                color = self._get_theme_color("user_prompt", "cyan")
                if state['context_mode'] == self.mode_lines:
                    text = html.escape(f"Ask [Ctx: {state['context_lines']}/{state['total_lines']}L]: ")
                    return HTML(f'<style fg="{color}">{text}</style>')
                active = get_active_buffer()
                lines_count = len(active.split('\n'))
                mode_str = {self.mode_range: "Range", self.mode_single: "Cmd"}[state['context_mode']]
                text = html.escape(f"Ask [{mode_str} {state['context_cmd']} ~{lines_count}L]: ")
                return HTML(f'<style fg="{color}">{text}</style>')
            from prompt_toolkit.application.current import get_app
            def get_toolbar():
                import html
                app = get_app()
                c_warning = self._get_theme_color("warning", "yellow")
                if app and app.current_buffer:
                    text = app.current_buffer.text
                    # Solo mostrar ayuda de comandos si estamos escribiendo el primer comando y no hay espacios
                    if text.startswith('/') and ' ' not in text:
                        commands = ['/os', '/prompt', '/architect', '/engineer', '/trust', '/untrust', '/memorize', '/clear']
                        matches = [c for c in commands if c.startswith(text.lower())]
                        if matches:
                            m_text = html.escape(f"Available: {' '.join(matches)}")
                            return HTML(f'<style fg="{c_warning}">{m_text}</style>' + " " * 20)
                m_label = {self.mode_range: "RANGE", self.mode_single: "SINGLE", self.mode_lines: "LINES"}[state['context_mode']]
                if state['context_mode'] == self.mode_lines:
                    base_str = f'\u25b6 Ctrl+\u2191/\u2193 adjusts by 50 lines  [Tab: {m_label}]'
                else:
                    idx = max(0, state['total_cmds'] - state['context_cmd'])
                    def clean_preview(text):
                        # Limpia saltos de línea y el prompt inicial (todo hasta #, > o $) para que quede solo el comando
                        original = text.strip().replace('\r', '').replace('\n', ' ')
                        cleaned = re.sub(r'^.*?[#>\$]\s*', '', original)
                        # Si limpiar el prompt nos deja con un string vacío (ej: era solo "iol#"), devolvemos el original
                        return cleaned if cleaned else original
                    if state['context_mode'] == self.mode_range:
                        range_blocks = blocks[idx:]
                        # Si hay más de un bloque, el último es siempre el prompt vacío/actual. Lo omitimos visualmente.
                        if len(range_blocks) > 1:
                            range_blocks = range_blocks[:-1]
                        # Limpiar y truncar comandos muy largos para que no rompan la UI
                        previews = []
                        for b in range_blocks:
                            p = clean_preview(b[2])
                            if p:
                                # Truncar comandos individuales largos
                                if len(p) > 25: p = p[:22] + "..."
                                previews.append(p)
                        if not previews:
                            desc = clean_preview(blocks[idx][2])
                        elif len(previews) <= 3:
                            desc = " + ".join(previews)
                        else:
                            desc = f"{previews[0]} + {previews[1]} + {previews[2]} ... (+{len(previews)-3})"
                    else:
                        # Modo SINGLE original
                        desc = clean_preview(blocks[idx][2])
                    base_str = f'\u25b6 {desc}  [Tab: {m_label}]'
                # Wrap base_str in a style to maintain consistency and avoid glitches
                # The fg color will be inherited from bottom-toolbar global style if not specified here
                base_html = f'<span>{html.escape(base_str)}</span>'
                res_html = base_html
                if state.get('toolbar_msg'):
                    if time.time() < state.get('msg_expiry', 0):
                        msg = html.escape(state['toolbar_msg'])
                        res_html = f'<style fg="{c_warning}">⚙️ {msg}</style> | ' + base_html
                    else:
                        state['toolbar_msg'] = ''
                # Pad with spaces to ensure the line is cleared when the message disappears
                return HTML(res_html + " " * 20)
            from prompt_toolkit.completion import Completer, Completion
            class SlashCommandCompleter(Completer):
                def get_completions(self, document, complete_event):
                    text = document.text_before_cursor
                    if text.startswith('/'):
                        parts = text.split()
                        # Only autocomplete the first word
                        if len(parts) <= 1 or (len(parts) == 1 and not text.endswith(' ')):
                            cmd_part = parts[0] if parts else text
                            commands = [
                                ('/os', 'Set device OS (e.g. cisco_ios)'),
                                ('/prompt', 'Override prompt regex'),
                                ('/architect', 'Switch to Architect persona'),
                                ('/engineer', 'Switch to Engineer persona'),
                                ('/trust', 'Enable auto-execute'),
                                ('/untrust', 'Disable auto-execute'),
                                ('/memorize', 'Add fact to memory'),
                                ('/clear', 'Clear memory')
                            ]
                            for cmd, desc in commands:
                                if cmd.startswith(cmd_part.lower()):
                                    yield Completion(cmd, start_position=-len(cmd_part), display_meta=desc)
            copilot_completer = SlashCommandCompleter()
            while True:
                # 2. Ask question
                from prompt_toolkit.styles import Style
                c_contrast = self._get_theme_color("contrast", "gray")
                ui_style = Style.from_dict({
                    'bottom-toolbar': f'fg:{c_contrast}',
                })
                session = PromptSession(
                    history=self.history, 
                    input=self.pt_input, 
                    output=self.pt_output,
                    completer=copilot_completer,
                    reserve_space_for_menu=0,
                    style=ui_style
                )
                try:
                    # Usamos un try/finally interno para asegurar que si algo falla en prompt_async,
                    # no nos quedemos con la terminal en un estado extraño.
                    question = await session.prompt_async(
                        get_prompt_text, 
                        key_bindings=bindings, 
                        bottom_toolbar=get_toolbar
                    )
                except (KeyboardInterrupt, EOFError):
                    state['cancelled'] = True
                    question = ""
                if state['cancelled'] or not question.strip() or question.strip().lower() in ['cancel', 'exit', 'quit']:
                    return "cancel", None, None
                # 3. Process Input via AIService
                directive = self.ai_service.process_copilot_input(question, self.session_state)
                if directive["action"] == "state_update":
                    state['toolbar_msg'] = directive['message']
                    state['msg_expiry'] = time.time() + 3 # 3 seconds timeout
                    async def delayed_refresh():
                        await asyncio.sleep(3.1)
                        # Only invalidate if the message hasn't been replaced by a newer one
                        if state.get('toolbar_msg') == directive['message']:
                            state['toolbar_msg'] = '' # Explicitly clear
                            try:
                                from prompt_toolkit.application.current import get_app
                                app = get_app()
                                if app: app.invalidate()
                            except: pass
                    asyncio.create_task(delayed_refresh())
                    # Mover el cursor arriba y limpiar la línea para que el nuevo prompt reemplace al anterior
                    sys.stdout.write('\x1b[1A\x1b[2K')
                    sys.stdout.flush()
                    continue
                else:
                    # Limpiar el mensaje de la barra cuando se hace una pregunta real
                    state['toolbar_msg'] = ''
                clean_question = directive.get("clean_prompt", question)
                overrides = directive.get("overrides", {})
                # Merge node_info with session_state and overrides
                merged_node_info = node_info.copy()
                if self.session_state['os']: merged_node_info['os'] = self.session_state['os']
                if self.session_state['prompt']: merged_node_info['prompt'] = self.session_state['prompt']
                merged_node_info['persona'] = self.session_state['persona']
                merged_node_info['trust'] = self.session_state['trust_mode']
                merged_node_info['memories'] = list(self.session_state['memories'])
                for k, v in overrides.items():
                    merged_node_info[k] = v
                # Enrich question
                past = self.history.get_strings()
                if len(past) > 1:
                    clean_past = [q for q in past[-6:-1] if not q.startswith('/')]
                    if clean_past:
                        history_text = "\n".join(f"- {q}" for q in clean_past)
                        clean_question = f"Previous questions:\n{history_text}\n\nCurrent Question:\n{clean_question}"
                # 3. AI Execution
                # Use persona from overrides (one-shot) or from session state
                active_persona = merged_node_info.get('persona', self.session_state.get('persona', 'engineer'))
                persona_color = self._get_theme_color(active_persona, fallback="cyan")
                persona_title = "Network Architect" if active_persona == "architect" else "Network Engineer"
                active_buffer = get_active_buffer()
                live_text = ""
                first_chunk = True
                from rich.rule import Rule
                from rich.status import Status
                from connpy.printer import IncrementalMarkdownParser
                md_parser = IncrementalMarkdownParser(console=self.console)
                status_spinner = Status(
                    f"[bold {persona_color}]{persona_title}:[/bold {persona_color}] [dim]Thinking...[/dim]",
                    console=self.console,
                    spinner="dots"
                )
                status_spinner.start()
                def on_chunk(text):
                    nonlocal live_text, first_chunk
                    if first_chunk:
                        status_spinner.stop()
                        # Print header rule before first chunk arrives
                        self.console.print(Rule(
                            f"[bold {persona_color}]{persona_title}[/bold {persona_color}]",
                            style=persona_color
                        ))
                        first_chunk = False
                    live_text += text
                    md_parser.feed(text)
                # Check for interruption during AI call
                ai_task = asyncio.create_task(on_ai_call(active_buffer, clean_question, on_chunk, merged_node_info))
                try:
                    while not ai_task.done():
                        await asyncio.sleep(0.05)
                    result = await ai_task
                except asyncio.CancelledError:
                    status_spinner.stop()
                    return "cancel", None, None
                # Ensure spinner is stopped if no chunks arrived
                if first_chunk:
                    status_spinner.stop()
                # Close the streamed output with a Rule
                if not first_chunk:
                    md_parser.flush()
                    self.console.print(Rule(style=persona_color))
                if not result or result.get("error"):
                    if first_chunk and result and result.get("error"):
                        self.console.print(f"[red]Error: {result['error']}[/red]")
                    return "cancel", None, None
                # If no chunks were streamed but we have a guide, print it as a panel
                if first_chunk and result and result.get("guide"):
                    self.console.print(Panel(Markdown(result["guide"]), title=f"[bold {persona_color}]{persona_title}[/bold {persona_color}]", border_style=persona_color))
                commands = result.get("commands", [])
                if not commands:
                    self.console.print("")
                    return "continue", None, None
                risk = result.get("risk_level", "low")
                risk_style = {"low": "success", "high": "warning", "destructive": "error"}.get(risk, "success")
                style_color = self._get_theme_color(risk_style, fallback="green")
                cmd_text = "\n".join(f"  {i+1}. {c}" for i, c in enumerate(commands))
                # Explicitly use 'bold style_color' for both TITLE and BORDER to ensure maximum consistency
                self.console.print(Panel(cmd_text, title=f"[bold {style_color}]Suggested Commands [{risk.upper()}][/bold {style_color}]", border_style=f"bold {style_color}"))
                if merged_node_info.get('trust', False) and risk != "destructive":
                    self.console.print(f"[dim]⚙️ Auto-executing (Trust Mode)[/dim]")
                    return "send_all", commands, None
                confirm_session = PromptSession(input=self.pt_input, output=self.pt_output)
                c_bindings = KeyBindings()
                @c_bindings.add('escape', eager=True)
                @c_bindings.add('c-c')
                def _(ev): ev.app.exit(result='n')
                import html
                try:
                    p_text = html.escape(f"Send? (y/n/e/range) [n]: ")
                    # Use the EXACT same style_color and force bold="true" for Prompt-Toolkit
                    action = await confirm_session.prompt_async(HTML(f'<style fg="{style_color}" bold="true">{p_text}</style>'), key_bindings=c_bindings)
                except (KeyboardInterrupt, EOFError):
                    self.console.print("")
                    return "continue", None, None
                def parse_indices(text, max_len):
                    """Helper to parse '1-3, 5, 7' into [0, 1, 2, 4, 6]."""
                    indices = []
                    # Replace commas with spaces and split
                    parts = text.replace(',', ' ').split()
                    for part in parts:
                        if '-' in part:
                            try:
                                start, end = map(int, part.split('-'))
                                # Ensure inclusive and 0-indexed
                                indices.extend(range(start-1, end))
                            except: continue
                        elif part.isdigit():
                            indices.append(int(part)-1)
                    # Filter valid indices and remove duplicates
                    return [i for i in sorted(set(indices)) if 0 <= i < max_len]
                action_l = (action or "n").lower().strip()
                if action_l in ('y', 'yes', 'all'):
                    return "send_all", commands, None
                # Check for numeric selection (e.g., "1, 2-4")
                if re.match(r'^[0-9,\-\s]+$', action_l):
                    selected_idxs = parse_indices(action_l, len(commands))
                    if selected_idxs:
                        return "send_all", [commands[i] for i in selected_idxs], None
                elif action_l.startswith('e'):
                    # Check if it's a selective edit like 'e1-2'
                    selection_str = action_l[1:].strip()
                    if selection_str:
                        idxs = parse_indices(selection_str, len(commands))
                        cmds_to_edit = [commands[i] for i in idxs] if idxs else commands
                    else:
                        cmds_to_edit = commands
                    target = "\n".join(cmds_to_edit)
                    e_bindings = KeyBindings()
                    @e_bindings.add('c-j')
                    def _(ev): ev.app.exit(result=ev.app.current_buffer.text)
                    @e_bindings.add('escape', 'enter')
                    def _(ev): ev.app.exit(result=ev.app.current_buffer.text)
                    @e_bindings.add('escape')
                    def _(ev): ev.app.exit(result='')
                    c_edit = self._get_theme_color("user_prompt", "cyan")
                    import html
                    e_text = html.escape("Edit (Ctrl+Enter or Esc+Enter to submit):\n")
                    try:
                        edited = await confirm_session.prompt_async(
                            HTML(f'<style fg="{c_edit}">{e_text}</style>'),
                            default=target, multiline=True, key_bindings=e_bindings
                        )
                    except (KeyboardInterrupt, EOFError):
                        self.console.print("")
                        return "continue", None, None
                    if edited and edited.strip():
                        # Split by lines to ensure core.py applies delay between each command
                        lines = [l.strip() for l in edited.split('\n') if l.strip()]
                        return "custom", None, lines
                self.console.print("")
                return "continue", None, None
            return "cancel", None, None
        finally:
            state['cancelled'] = True
@@ -169,11 +169,21 @@ def _build_tree(nodes, folders, profiles, plugins, configdir):
    }
    # State Machine Definitions
    mcp_dict = {
        "list": None,
        "add": {"*": {"*": {"*": None}}}, # name url [os]
        "remove": {"*": None},
        "enable": {"*": None},
        "disable": {"*": None},
        "--help": None, "-h": None
    }
    ai_dict = {"__exclude_used__": True, "--help": None, "-h": None}
    for opt in ["--engineer-model", "--engineer-api-key", "--architect-model", "--architect-api-key"]:
        ai_dict[opt] = {"*": ai_dict} # takes value, loops back
    for opt in ["--debug", "--trust", "--list", "--list-sessions", "--session", "--resume", "--delete", "--delete-session", "-y"]:
        ai_dict[opt] = ai_dict # takes no value, loops back
    ai_dict["--mcp"] = mcp_dict
    ai_dict["*"] = ai_dict
    mv_state = {"__extra__": _nodes, "--help": None, "-h": None}
@@ -184,9 +194,37 @@ def _build_tree(nodes, folders, profiles, plugins, configdir):
        "folders": None,
    }
    # --- Connect (default command) ---
    # Long flags are offered; short forms (-d/-t) only used for navigation.
    # Two states: before node (offer nodes + remaining long flags)
    #             after node  (offer only remaining long flags, no more nodes)
    connect_flags_long = ["--debug", "--sftp"]
    connect_flags_all  = ["--debug", "-d", "--sftp", "-t"]
    # Post-node: only offer remaining long flags
    connect_after_node = {"__exclude_used__": True}
    for f in connect_flags_all:
        connect_after_node[f] = connect_after_node
    # Pre-node: offer nodes + remaining long flags, consume node → post-node state
    connect_dict = {"__exclude_used__": True}
    connect_dict["__extra__"] = lambda w: (
        list(nodes) + list(folders) + (list(plugins.keys()) if plugins else [])
    )
    connect_dict["*"] = connect_after_node
    for f in connect_flags_all:
        connect_dict[f] = connect_dict
    # --- Main Tree ---
    return {
        # Root: offer nodes + long flags; after a node go to post-node state
        "__extra__": lambda w: list(nodes) + list(folders) + (list(plugins.keys()) if plugins else []),
        "*": connect_after_node,
        "--debug": connect_dict,
        "-d":      connect_dict,
        "--sftp":  connect_dict,
        "-t":      connect_dict,
        "--add": {"profile": _profile_values},
        "--del": {"profile": _profile_values, "__extra__": _nodes_folders},
@@ -89,6 +89,10 @@ class connapp:
        if hasattr(self.services.nodes, "list_folders") and hasattr(self.services.nodes.list_folders, "register_post_hook"):
            self.services.nodes.list_folders.register_post_hook(self.services.context.filter_node_list)
        # Apply theme from config if exists before remote connection attempts
        user_theme = self.config.config.get("theme", {})
        self._apply_app_theme(user_theme)
        # Populate data via services
        try:
            self.nodes_list = self.services.nodes.list_nodes()
@@ -152,10 +156,6 @@ class connapp:
        configfile._saveconfig.register_post_hook(auto_sync_hook)
        # Apply theme from config if exists
        user_theme = self.config.config.get("theme", {})
        self._apply_app_theme(user_theme)
    def _apply_app_theme(self, styles):
        """Unified method to apply theme to printer and help formatter."""
        active_styles = printer.apply_theme(styles)
@@ -284,6 +284,7 @@ class connapp:
        aiparser.add_argument("--session", nargs=1, help="Resume a specific AI session by ID")
        aiparser.add_argument("--resume", action="store_true", help="Resume the most recent AI session")
        aiparser.add_argument("--delete", "--delete-session", dest="delete_session", nargs=1, help="Delete an AI session by ID")
        aiparser.add_argument("--mcp", nargs='*', metavar=('ACTION', 'NAME'), help="Manage MCP servers. Actions: list, add, remove, enable, disable. Leave empty for interactive wizard.")
        aiparser.set_defaults(func=self._ai.dispatch)
        #RUNPARSER
        runparser = subparsers.add_parser("run", help="Run scripts or commands on nodes", description="Run scripts or commands on nodes", formatter_class=RichHelpFormatter) 
@@ -470,6 +471,13 @@ class connapp:
            # Handle global Ctrl+C gracefully
            printer.warning("Operation cancelled by user.")
            sys.exit(130)
        finally:
            # Safely cleanup AI sessions (litellm)
            try:
                from .ai import cleanup
                cleanup()
            except ImportError:
                pass
    class _store_type(argparse.Action):
        #Custom store type for cli app.
@@ -18,7 +18,29 @@ import asyncio
 import fcntl
 from . import printer
 from .tunnels import LocalStream
 from contextlib import contextmanager
@contextmanager
 def copilot_terminal_mode():
    import sys, tty, termios
    fd = sys.stdin.fileno()
    try:
        old_settings = termios.tcgetattr(fd)
        # Primero pasamos a raw mode absoluto para matar ISIG, ICANON, ECHO, etc.
        tty.setraw(fd)
        # Luego rehabilitamos OPOST para que rich.Live se dibuje correctamente
        new_settings = termios.tcgetattr(fd)
        new_settings[1] = new_settings[1] | termios.OPOST
        termios.tcsetattr(fd, termios.TCSANOW, new_settings)
        yield
    finally:
        try:
            termios.tcsetattr(fd, termios.TCSANOW, old_settings)
        except Exception:
            pass
 #functions and classes
@ClassHook
@@ -75,6 +97,7 @@ class node:
            - jumphost (str): Reference another node to be used as a jumphost
        '''
        self.config = config
        if config == '':
            self.idletime = 0
            self.key = None
@@ -188,6 +211,7 @@ class node:
        self.output = ""
        self.status = 1
        self.result = {}
        self.cmd_byte_positions = [(0, None)]
    @MethodHook
    def _passtx(self, passwords, *, keyfile=None):
@@ -232,61 +256,29 @@ class node:
    @MethodHook
    def _logclean(self, logfile, var = False):
-        # Remove special ascii characters and process terminal cursor movements to clean logs.
+        """Remove special ascii characters and process terminal cursor movements to clean logs."""
        from .utils import log_cleaner
        if var == False:
-            t = open(logfile, "r").read()
+            try:
                with open(logfile, "r") as f:
                    t = f.read()
            except:
                return
        else:
            t = logfile
-        lines = t.split('\n')
+        result = log_cleaner(t)
        cleaned_lines = []
        # Regex to capture: ANSI sequences, control characters (\r, \b, etc), and plain text chunks
        token_re = re.compile(r'(\x1B(?:[@-Z\\-_]|\[[0-?]*[ -/ ]*[@-~])|\r|\b|\x7f|[\x00-\x1F]|[^\x1B\r\b\x7f\x00-\x1F]+)')
        for line in lines:
            buffer = []
            cursor = 0
            for token in token_re.findall(line):
                if token == '\r':
                    cursor = 0
                elif token in ('\b', '\x7f'):
                    if cursor > 0:
                        cursor -= 1
                elif token == '\x1B[D': # Left Arrow
                    if cursor > 0:
                        cursor -= 1
                elif token == '\x1B[C': # Right Arrow
                    if cursor < len(buffer):
                        cursor += 1
                elif token == '\x1B[K': # Clear to end of line
                    buffer = buffer[:cursor]
                elif token.startswith('\x1B'):
                    # Ignore other ANSI sequences (colors, etc)
                    continue
                elif len(token) == 1 and ord(token) < 32:
                    # Ignore other non-printable control chars
                    continue
                else:
                    # Regular printable text
                    for char in token:
                        if cursor == len(buffer):
                            buffer.append(char)
                        else:
                            buffer[cursor] = char
                        cursor += 1
            cleaned_lines.append("".join(buffer))
        t = "\n".join(cleaned_lines).replace('\n\n', '\n').strip()
        if var == False:
-            d = open(logfile, "w")
+            try:
-            d.write(t)
+                with open(logfile, "w") as f:
-            d.close()
+                    f.write(result)
            except:
                pass
            return
        else:
-            return t
+            return result
    @MethodHook
    def _savelog(self):
@@ -329,14 +321,17 @@ class node:
            port_str = f":{self.port}" if self.port and self.protocol not in ["ssm", "kubectl", "docker"] else ""
            logger("success", f"Connected to {self.unique} at {self.host}{port_str} via: {self.protocol}")
-        if 'logfile' in dir(self):
+        # Always initialize self.mylog to capture terminal context for the AI Copilot
-            # Initialize self.mylog
+        if not hasattr(self, 'mylog'):
            if not 'mylog' in dir(self):
            self.mylog = io.BytesIO()
        if not async_mode:
            self.child.logfile_read = self.mylog
-                # Start the _savelog thread
+        # Only start disk-logging tasks if logfile is configured
        if 'logfile' in dir(self):
            if not async_mode:
                # Start the _savelog thread (sync mode)
                log_thread = threading.Thread(target=self._savelog)
                log_thread.daemon = True
                log_thread.start()
@@ -348,6 +343,7 @@ class node:
            x.start()
        if debug:
            if 'mylog' in dir(self):
                if not async_mode:
                    print(self.mylog.getvalue().decode())
    def _teardown_interact_environment(self):
@@ -355,7 +351,7 @@ class node:
            with open(self.logfile, "w") as f:
                f.write(self._logclean(self.mylog.getvalue().decode(), True))
-    async def _async_interact_loop(self, local_stream, resize_callback):
+    async def _async_interact_loop(self, local_stream, resize_callback, copilot_handler=None):
        local_stream.setup(resize_callback=resize_callback)
        try:
            child_fd = self.child.child_fd
@@ -390,9 +386,14 @@ class node:
            loop = asyncio.get_running_loop()
            child_reader_queue = asyncio.Queue()
            # Reset and track command byte positions for copilot context navigation
            # Each entry is (byte_position, command_text_or_None)
            self.cmd_byte_positions = [(self.mylog.tell() if hasattr(self, 'mylog') else 0, None)]
            def _child_read_ready():
                try:
-                    data = os.read(child_fd, 4096)
+                    # Increase buffer to 64KB for better high-speed handling
                    data = os.read(child_fd, 65536)
                    if data:
                        child_reader_queue.put_nowait(data)
                    else:
@@ -410,8 +411,35 @@ class node:
                    data = await local_stream.read()
                    if not data:
                        break
-                    try:
+                    
-                        os.write(child_fd, data)
+                    # Copilot interception
                    if copilot_handler and b'\x00' in data:
                        # Build node info from available metadata and ensure values are strings (not bytes)
                        def to_str(val):
                            if isinstance(val, bytes):
                                return val.decode(errors='replace')
                            return str(val) if val is not None else "unknown"
                        node_info = {
                            "name": to_str(getattr(self, 'unique', 'unknown')), 
                            "host": to_str(getattr(self, 'host', 'unknown'))
                        }
                        if isinstance(getattr(self, 'tags', None), dict):
                            node_info["os"] = to_str(self.tags.get("os", "unknown"))
                            node_info["prompt"] = to_str(self.tags.get("prompt", r'>$|#$|\$$|>.$|#.$|\$.$'))
                        # Invoke copilot (async callback handles UI)
                        await copilot_handler(self.mylog.getvalue(), node_info, local_stream, child_fd, self.cmd_byte_positions)
                        continue
                    # Remove any stray \x00 bytes and forward normally
                    clean_data = data.replace(b'\x00', b'')
                    if clean_data:
                        # Track command boundaries when user hits Enter
                        if hasattr(self, 'mylog') and (b'\r' in clean_data or b'\n' in clean_data):
                            self.cmd_byte_positions.append((self.mylog.tell(), None))
                        try:                            os.write(child_fd, clean_data)
                        except OSError:
                            break
                        self.lastinput = time()
@@ -424,17 +452,40 @@ class node:
                    if not data:
                        break
                    # Batching Optimization: Drain the queue to batch writes during high-volume bursts
                    # Helps the terminal parse ANSI faster and reduces syscalls.
                    chunks = [data]
                    while not child_reader_queue.empty():
                        try:
                            extra = child_reader_queue.get_nowait()
                            if not extra:
                                chunks.append(b'') # Re-put EOF later or handle it
                                break
                            chunks.append(extra)
                        except asyncio.QueueEmpty:
                            break
                    has_eof = chunks[-1] == b''
                    if has_eof:
                        chunks.pop()
                    if chunks:
                        combined_data = b''.join(chunks)
                        if skip_newlines:
-                        stripped = data.lstrip(b'\r\n')
+                            stripped = combined_data.lstrip(b'\r\n')
                            if stripped:
                                skip_newlines = False
-                            data = stripped
+                                combined_data = stripped
                            else:
                                if has_eof: break
                                continue
-                    await local_stream.write(data)
+                        await local_stream.write(combined_data)
                        if hasattr(self, 'mylog'):
-                        self.mylog.write(data)
+                            self.mylog.write(combined_data)
                    if has_eof:
                        break
            async def keepalive_task():
                while True:
@@ -453,16 +504,17 @@ class node:
                    current_size = self.mylog.tell()
                    if current_size != prev_size:
                        try:
                            # Move heavy log cleaning to a thread to avoid freezing the interaction loop
                            raw_log = self.mylog.getvalue().decode(errors='replace')
                            cleaned_log = await asyncio.to_thread(self._logclean, raw_log, True)
                            with open(self.logfile, "w") as f:
-                                f.write(self._logclean(self.mylog.getvalue().decode(), True))
+                                f.write(cleaned_log)
                            prev_size = current_size
                        except Exception:
                            pass
            try:
-                # gather runs until any task completes (or we just let them run until EOF breaks them)
+                # We wait for either the user (ingress) or the child (egress) to finish
                # Ingress breaks on user EOF. Egress breaks on child EOF. 
                # We want to exit if either happens, so return_exceptions=False, but we need to cancel the others.
                tasks = [
                    asyncio.create_task(ingress_task()),
                    asyncio.create_task(egress_task())
@@ -471,9 +523,34 @@ class node:
                    tasks.append(asyncio.create_task(keepalive_task()))
                if hasattr(self, 'logfile') and hasattr(self, 'mylog'):
                    tasks.append(asyncio.create_task(savelog_task()))
-                done, pending = await asyncio.wait(tasks, return_when=asyncio.FIRST_COMPLETED)
+                
-                for p in pending:
+                done, pending = await asyncio.wait(
-                    p.cancel()
+                    [tasks[0], tasks[1]], 
                    return_when=asyncio.FIRST_COMPLETED
                )
                # If ingress finished first (user quit), give egress a small window to catch up 
                # on the remaining output in the queue.
                if tasks[0] in done and tasks[1] not in done:
                    try:
                        await asyncio.wait_for(tasks[1], timeout=0.2)
                    except (asyncio.TimeoutError, asyncio.CancelledError):
                        pass
                for t in tasks:
                    if t not in done:
                        t.cancel()
                # Final log sync on thread to avoid losing last lines
                if hasattr(self, 'logfile') and hasattr(self, 'mylog'):
                    try:
                        raw_log = self.mylog.getvalue().decode(errors='replace')
                        cleaned_log = await asyncio.to_thread(self._logclean, raw_log, True)
                        with open(self.logfile, "w") as f:
                            f.write(cleaned_log)
                    except Exception:
                        pass
            finally:
                loop.remove_reader(child_fd)
                try:
@@ -484,6 +561,45 @@ class node:
        finally:
            local_stream.teardown()
    @MethodHook
    async def inject_commands(self, commands, child_fd, on_inject=None):
        """
        Inject a list of commands into the node's PTY.
        Handles screen_length_command, history tracking and delays.
        """
        if not commands:
            return
        # 0. Clear line
        os.write(child_fd, b'\x15')
        await asyncio.sleep(0.1)
        # 1. Prepare list (prepend screen_length if exists)
        slc = self.tags.get("screen_length_command") if hasattr(self, 'tags') and isinstance(self.tags, dict) else None
        to_send = list(commands)
        if slc and slc not in to_send: # avoid duplicates if already there
             to_send.insert(0, slc)
        # 2. Inject one by one
        for cmd in to_send:
            # Register in node's official history (SKIP if it's the administrative screen length command)
            if cmd != slc and hasattr(self, 'cmd_byte_positions') and self.cmd_byte_positions is not None:
                log_pos = self.mylog.tell() if hasattr(self, 'mylog') else 0
                self.cmd_byte_positions.append((log_pos, cmd))
            # Write physically to PTY
            os.write(child_fd, (cmd + "\n").encode())
            # Notify (e.g., for gRPC or logs) - SKIP for administrative SLC
            if on_inject and cmd != slc:
                if asyncio.iscoroutinefunction(on_inject):
                    await on_inject(cmd)
                else:
                    on_inject(cmd)
            # Delay to avoid overwhelming the router
            await asyncio.sleep(0.8)
    @MethodHook
    def interact(self, debug=False, logger=None):
@@ -504,7 +620,10 @@ class node:
                    except Exception:
                        pass
-                asyncio.run(self._async_interact_loop(local_stream, resize_callback))
+                # Build local copilot handler
                copilot_handler = self._build_local_copilot_handler()
                asyncio.run(self._async_interact_loop(local_stream, resize_callback, copilot_handler=copilot_handler))
            finally:
                self._teardown_interact_environment()
        else:
@@ -514,6 +633,81 @@ class node:
                printer.error(f"Connection failed: {str(connect)}")
            sys.exit(1)
    def _build_local_copilot_handler(self):
        """Build copilot handler for local CLI sessions using rich for rendering."""
        config = getattr(self, 'config', None) if hasattr(self, 'config') else None
        return self._copilot_handler(config)
    def _copilot_handler(self, config):
        """Unified copilot handler for local session."""
        from .cli.terminal_ui import CopilotInterface
        from .services.ai_service import AIService
        import asyncio
        import os
        async def handler(buffer, node_info, stream, child_fd, cmd_byte_positions=None):
            try:
                interface = CopilotInterface(
                    config, 
                    history=getattr(stream, 'copilot_history', None),
                    session_state=getattr(stream, 'copilot_state', None)
                )
                # Save history back to stream for persistence in current session
                stream.copilot_history = interface.history
                stream.copilot_state = interface.session_state
                ai_service = AIService(config)
                async def on_ai_call(active_buffer, question, chunk_callback, merged_node_info):
                    return await ai_service.aask_copilot(
                        active_buffer,
                        question,
                        node_info=merged_node_info,
                        chunk_callback=chunk_callback
                    )
                # Get raw bytes from BytesIO
                raw_bytes = self.mylog.getvalue()
                # Detener el lector de la terminal para que prompt_toolkit (en run_session) 
                # tenga control exclusivo del stdin sin interferencias de LocalStream.
                if hasattr(stream, 'stop_reading'):
                    stream.stop_reading()
                elif hasattr(stream, '_loop') and hasattr(stream, 'stdin_fd'):
                    # Fallback si no tiene el método (en LocalStream)
                    stream._loop.remove_reader(stream.stdin_fd)
                try:
                    with copilot_terminal_mode():
                        while True:
                            action, commands, custom_cmd = await interface.run_session(
                                raw_bytes=raw_bytes,
                                cmd_byte_positions=self.cmd_byte_positions,
                                node_info=node_info,
                                on_ai_call=on_ai_call
                            )
                            if action == "continue":
                                continue
                            break
                finally:
                    print("\033[2m Returning to session...\033[0m", flush=True)
                    # Reiniciar el lector de la terminal para volver al modo interactivo SSH/Telnet
                    if hasattr(stream, 'start_reading'):
                        stream.start_reading()
                    elif hasattr(stream, '_loop') and hasattr(stream, 'stdin_fd'):
                        stream._loop.add_reader(stream.stdin_fd, stream._read_ready)
                if action in ("send_all", "custom"):
                    cmds_to_send = commands if action == "send_all" else custom_cmd
                    await self.inject_commands(cmds_to_send, child_fd)
                else:
                    os.write(child_fd, b'\x15\r')
            except Exception as e:
                import traceback
                print(f"\n[ERROR in Copilot Handler] {e}", flush=True)
                traceback.print_exc()
                os.write(child_fd, b'\x15\r')
        return handler
    @MethodHook
    def run(self, commands, vars = None,*, folder = '', prompt = r'>$|#$|\$$|>.$|#.$|\$.$', stdout = False, timeout = 10, logger = None):
@@ -575,7 +769,6 @@ class node:
            if "prompt" in self.tags:
                prompt = self.tags["prompt"]
            expects = [prompt, pexpect.EOF, pexpect.TIMEOUT]
            output = ''
            status = ''
            if not isinstance(commands, list):
@@ -686,7 +879,6 @@ class node:
            if "prompt" in self.tags:
                prompt = self.tags["prompt"]
            expects = [prompt, pexpect.EOF, pexpect.TIMEOUT]
            output = ''
            if not isinstance(commands, list):
                commands = [commands]
@@ -752,15 +944,18 @@ class node:
    @MethodHook
    def _generate_ssh_sftp_cmd(self):
        cmd = self.protocol
        if self.idletime > 0:
            cmd += " -o ServerAliveInterval=" + str(self.idletime)
        if self.port:
            if self.protocol == "ssh":
                cmd += " -p " + self.port
            elif self.protocol == "sftp":
                cmd += " -P " + self.port
        if self.options:
-            cmd += " " + self.options
+            opts = self.options
            if self.protocol == "sftp":
                # Strip SSH-only flags that sftp doesn't support
                opts = re.sub(r'(?<!\S)-[XxtTAaNf]\b', '', opts).strip()
            if opts:
                cmd += " " + opts
        if self.jumphost:
            cmd += " " + self.jumphost
        user_host = f"{self.user}@{self.host}" if self.user else self.host
@@ -803,6 +998,19 @@ class node:
            cmd += f" {self.options}"
        return cmd
    @MethodHook
    def _generate_ssm_cmd(self):
        region = self.tags.get("region", "") if isinstance(self.tags, dict) else ""
        profile = self.tags.get("profile", "") if isinstance(self.tags, dict) else ""
        cmd = f"aws ssm start-session --target {self.host}"
        if region:
            cmd += f" --region {region}"
        if profile:
            cmd += f" --profile {profile}"
        if self.options:
            cmd += f" {self.options}"
        return cmd
    @MethodHook
    def _get_cmd(self):
        if self.protocol in ["ssh", "sftp"]:
@@ -875,6 +1083,7 @@ class node:
                if logger:
                    logger("debug", f"Command:\n{cmd}")
                self.mylog = io.BytesIO()
                self.mylog.write(f"[i] [DEBUG] Command:\r\n    {cmd}\r\n".encode())
                child.logfile_read = self.mylog
@@ -55,8 +55,13 @@ def handle_errors(func):
        return wrapper
 class NodeServicer(connpy_pb2_grpc.NodeServiceServicer):
-    def __init__(self, config):
+    def __init__(self, config, debug=False):
        self.service = NodeService(config)
        self.server_debug = debug
        if debug:
            from rich.console import Console
            from ..printer import connpy_theme, get_original_stdout
            self.server_console = Console(theme=connpy_theme, file=get_original_stdout())
    @handle_errors
    def interact_node(self, request_iterator, context):
@@ -79,8 +84,8 @@ class NodeServicer(connpy_pb2_grpc.NodeServiceServicer):
        sftp = first_req.sftp
        debug = first_req.debug
-        if debug:
+        if self.server_debug:
-            printer.console.print(f"[debug][DEBUG][/debug] gRPC interact_node request for: [bold cyan]{unique_id}[/bold cyan]")
+            self.server_console.print(f"[debug][DEBUG][/debug] gRPC interact_node request for: [bold cyan]{unique_id}[/bold cyan]")
        if first_req.connection_params_json:
            import json
@@ -139,7 +144,39 @@ class NodeServicer(connpy_pb2_grpc.NodeServiceServicer):
            if sftp:
                n.protocol = "sftp"
-        connect = n._connect(debug=debug)
+        # Build a logger that captures debug messages as ANSI-colored bytes for the client
        debug_chunks = []
        if debug:
            from io import StringIO
            from rich.console import Console as RichConsole
            from ..printer import connpy_theme
            from .. import printer as _printer
            def remote_logger(msg_type, message):
                buf = StringIO()
                c = RichConsole(file=buf, force_terminal=True, width=120, theme=connpy_theme)
                if msg_type == "debug":
                    c.print(_printer._format_multiline("i", f"[DEBUG] {message}", style="info"))
                elif msg_type == "success":
                    c.print(_printer._format_multiline("✓", message, style="success"))
                elif msg_type == "error":
                    c.print(_printer._format_multiline("✗", message, style="error"))
                else:
                    c.print(str(message))
                rendered = buf.getvalue()
                if rendered:
                    # Raw TTY needs \r\n instead of \n
                    rendered = rendered.replace('\n', '\r\n')
                    debug_chunks.append(rendered.encode())
        else:
            remote_logger = None
        connect = n._connect(debug=debug, logger=remote_logger)
        # Send debug output to client before checking result (always show the command)
        for chunk in debug_chunks:
            yield connpy_pb2.InteractResponse(stdout_data=chunk)
        if connect != True:
            yield connpy_pb2.InteractResponse(success=False, error_message=str(connect))
            return
@@ -166,7 +203,177 @@ class NodeServicer(connpy_pb2_grpc.NodeServiceServicer):
                    except Exception:
                        pass
-                asyncio.run(n._async_interact_loop(remote_stream, resize_callback))
+                async def remote_copilot_handler(buffer, node_info, stream, child_fd, cmd_byte_positions=None):
                    import json
                    import asyncio
                    import os
                    from ..services.ai_service import AIService
                    service = AIService(self.service.config)
                    if node_info is None:
                        node_info = {}
                    # Calculate real command blocks from history using the central service
                    raw_bytes = n.mylog.getvalue() if hasattr(n, 'mylog') else buffer
                    if not isinstance(raw_bytes, bytes):
                        raw_bytes = str(raw_bytes).encode()
                    from connpy.utils import log_cleaner
                    last_line = log_cleaner(raw_bytes.decode(errors='replace')).split('\n')[-1].strip()
                    blocks = service.build_context_blocks(raw_bytes, n.cmd_byte_positions, node_info, last_line=last_line)
                    node_info["context_blocks"] = blocks
                    node_info_json = json.dumps(node_info)
                    # Convert buffer to string if it's bytes for the preview
                    preview_str = buffer[-200:].decode(errors='replace') if isinstance(buffer, bytes) else str(buffer)[-200:]
                    # Generate a unique session ID for this copilot interaction to prevent race conditions
                    import uuid
                    copilot_session_id = str(uuid.uuid4())
                    node_info["session_id"] = copilot_session_id
                    node_info_json = json.dumps(node_info)
                    # 1. Send prompt to client
                    response_queue.put(connpy_pb2.InteractResponse(
                        copilot_prompt=True,
                        copilot_buffer_preview=preview_str,
                        copilot_node_info_json=node_info_json
                    ))
                    while True:
                        # 0. Drain the queue of any stale messages before starting a new interaction
                        while not remote_stream.copilot_queue.empty():
                            try:
                                remote_stream.copilot_queue.get_nowait()
                            except:
                                break
                        # 2. Await the question from client via the copilot_queue
                        import threading
                        def preload_ai_deps():
                            try:
                                import litellm
                            except Exception:
                                pass
                        threading.Thread(target=preload_ai_deps, daemon=True).start()
                        try:
                            req_data = await asyncio.wait_for(remote_stream.copilot_queue.get(), timeout=120)
                            if not req_data: return
                            # Validate session ID if provided by client (skip validation if not provided for CLI compatibility)
                            req_session_id = req_data.get("session_id")
                            if req_session_id and req_session_id != copilot_session_id:
                                continue # Ignore stale request from a previous session
                            if "question" not in req_data or not req_data["question"] or req_data["question"] == "CANCEL" or req_data.get("action") in ("cancel", "web_cancel"):
                                if req_data.get("action") == "web_cancel":
                                    os.write(child_fd, b'\x05')
                                else:
                                    os.write(child_fd, b'\x15\r')
                                return
                            question = req_data["question"]
                            merged_node_info_str = req_data.get("node_info_json", "")
                            if merged_node_info_str:
                                try:
                                    merged_node_info = json.loads(merged_node_info_str)
                                    node_info.update(merged_node_info)
                                except: pass
                            context_buffer = req_data.get("context_buffer", "")
                            if context_buffer.startswith('{"context_start_pos"'):
                                try:
                                    parsed = json.loads(context_buffer)
                                    start_pos = parsed["context_start_pos"]
                                    selected_raw = raw_bytes[start_pos:]
                                    context_buffer = n._logclean(selected_raw.decode(errors='replace'), var=True)
                                except Exception:
                                    context_buffer = buffer
                            elif not context_buffer:
                                context_buffer = buffer
                        except asyncio.TimeoutError:
                            os.write(child_fd, b'\x15\r')
                            return
                        # 3. Call AI Service with streaming
                        def chunk_callback(chunk_text):
                            if chunk_text:
                                response_queue.put(connpy_pb2.InteractResponse(
                                    copilot_stream_chunk=chunk_text
                                ))
                        # Create a clean version of node_info for the AI to save tokens and match local CLI behavior
                        ai_node_info = {k: v for k, v in node_info.items() if k not in ("context_blocks", "full_buffer")}
                        ai_task = asyncio.create_task(service.aask_copilot(context_buffer, question, ai_node_info, chunk_callback=chunk_callback))
                        wait_action_task = asyncio.create_task(remote_stream.copilot_queue.get())
                        done, pending = await asyncio.wait(
                            [ai_task, wait_action_task],
                            return_when=asyncio.FIRST_COMPLETED
                        )
                        if wait_action_task in done:
                            req_data = wait_action_task.result()
                            ai_task.cancel()
                            if req_data.get("action") in ("cancel", "web_cancel") or req_data.get("question") == "CANCEL":
                                if req_data.get("action") == "web_cancel":
                                    os.write(child_fd, b'\x05')
                                else:
                                    os.write(child_fd, b'\x15\r')
                                return
                            continue # Loop back instead of returning to keep session alive
                        else:
                            wait_action_task.cancel()
                            result = ai_task.result()
                            if not result:
                                os.write(child_fd, b'\x15\r')
                                return
                        # 4. Send response back to client
                        response_queue.put(connpy_pb2.InteractResponse(
                            copilot_response_json=json.dumps(result)
                        ))
                        # 5. Wait for user action
                        try:
                            action_data = await asyncio.wait_for(remote_stream.copilot_queue.get(), timeout=60)
                            if not action_data: return
                            action = action_data.get("action", "cancel")
                            if action == "continue":
                                continue # Loop back for next question
                            if action in ("cancel", "web_cancel"):
                                if action == "web_cancel":
                                    os.write(child_fd, b'\x05')
                                else:
                                    os.write(child_fd, b'\x15\r')
                                return
                        except asyncio.TimeoutError:
                            os.write(child_fd, b'\x15\r')
                            return
                        def on_inject(cmd):
                            response_queue.put(connpy_pb2.InteractResponse(copilot_injected_command=cmd))
                        if action == "send_all":
                            commands = result.get("commands", [])
                            await n.inject_commands(commands, child_fd, on_inject=on_inject)
                            return
                        elif action.startswith("custom:"):
                            custom_cmds_raw = action[7:]
                            custom_cmds = [cmd.strip() for cmd in custom_cmds_raw.split('\n') if cmd.strip()]
                            await n.inject_commands(custom_cmds, child_fd, on_inject=on_inject)
                            return
                        else:
                            os.write(child_fd, b'\x15\r')
                            return
                asyncio.run(n._async_interact_loop(remote_stream, resize_callback, copilot_handler=remote_copilot_handler))
            except Exception as e:
                pass
            finally:
@@ -175,14 +382,19 @@ class NodeServicer(connpy_pb2_grpc.NodeServiceServicer):
        t_loop = threading.Thread(target=run_async_loop, daemon=True)
        t_loop.start()
-
+        def response_generator():
            while True:
                data = response_queue.get()
                if data is None:
-                if debug:
+                    if self.server_debug:
-                    printer.console.print(f"[debug][DEBUG][/debug] gRPC interact_node session closed for: [bold cyan]{unique_id}[/bold cyan]")
+                        self.server_console.print(f"[debug][DEBUG][/debug] gRPC interact_node session closed for: [bold cyan]{unique_id}[/bold cyan]")
                    break
                if isinstance(data, connpy_pb2.InteractResponse):
                    yield data
                else:
                    yield connpy_pb2.InteractResponse(stdout_data=data)
        yield from response_generator()
    @handle_errors
    def list_nodes(self, request, context):
        f = request.filter_str if request.filter_str else None
@@ -716,6 +928,22 @@ class AIServicer(connpy_pb2_grpc.AIServiceServicer):
        res = self.service.confirm(request.value)
        return connpy_pb2.BoolResponse(value=res)
    @handle_errors
    def ask_copilot(self, request, context):
        import json
        node_info = json.loads(request.node_info_json) if request.node_info_json else None
        result = self.service.ask_copilot(
            request.terminal_buffer,
            request.user_question,
            node_info
        )
        return connpy_pb2.CopilotResponse(
            commands=result.get("commands", []),
            guide=result.get("guide", ""),
            risk_level=result.get("risk_level", "low"),
            error=result.get("error") or ""
        )
    @handle_errors
    def list_sessions(self, request, context):
        return connpy_pb2.ValueResponse(data=to_value(self.service.list_sessions()))
@@ -730,6 +958,17 @@ class AIServicer(connpy_pb2_grpc.AIServiceServicer):
        self.service.configure_provider(request.provider, request.model, request.api_key)
        return Empty()
    @handle_errors
    def configure_mcp(self, request, context):
        self.service.configure_mcp(
            request.name, 
            url=request.url or None, 
            enabled=request.enabled, 
            auto_load_on_os=request.auto_load_on_os or None, 
            remove=request.remove
        )
        return Empty()
    @handle_errors
    def load_session_data(self, request, context):
        return connpy_pb2.StructResponse(data=to_struct(self.service.load_session_data(request.value)))
@@ -789,7 +1028,7 @@ def serve(config, port=8048, debug=False):
    interceptors = [LoggingInterceptor()] if debug else []
    server = grpc.server(futures.ThreadPoolExecutor(max_workers=10), interceptors=interceptors)
-    connpy_pb2_grpc.add_NodeServiceServicer_to_server(NodeServicer(config), server)
+    connpy_pb2_grpc.add_NodeServiceServicer_to_server(NodeServicer(config, debug=debug), server)
    connpy_pb2_grpc.add_ProfileServiceServicer_to_server(ProfileServicer(config), server)
    connpy_pb2_grpc.add_ConfigServiceServicer_to_server(ConfigServicer(config), server)
    plugin_servicer = PluginServicer(config)
@@ -9,6 +9,8 @@ from .utils import to_value, from_value, to_struct, from_struct
 from ..services.exceptions import ConnpyError
 from ..hooks import MethodHook
 from .. import printer
 from ..cli.terminal_ui import CopilotInterface
 from ..utils import log_cleaner
 def handle_errors(func):
    @wraps(func)
@@ -41,15 +43,100 @@ class NodeStub:
        self.remote_host = remote_host
        self.config = config
    def _handle_remote_copilot(self, res, request_queue, response_queue, client_buffer_bytes, pause_generator, resume_generator, old_tty):
        import json, asyncio, termios, sys, tty, queue
        from ..core import copilot_terminal_mode
        from . import connpy_pb2
        pause_generator()
        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
        node_info = json.loads(res.copilot_node_info_json) if res.copilot_node_info_json else {}
        blocks = node_info.get("context_blocks", [])
        interface = CopilotInterface(
            self.config, 
            history=getattr(self, 'copilot_history', None),
            session_state=getattr(self, 'copilot_state', None)
        )
        self.copilot_history = interface.history
        self.copilot_state = interface.session_state
        async def on_ai_call_remote(active_buffer, question, chunk_callback, merged_node_info):
            # Send request to server
            request_queue.put(connpy_pb2.InteractRequest(
                copilot_question=question, 
                copilot_context_buffer=active_buffer,
                copilot_node_info_json=json.dumps(merged_node_info)
            ))
            # Wait for chunks from server
            while True:
                try:
                    chunk_res = response_queue.get(timeout=0.1)
                    if chunk_res is None: return {"error": "Server disconnected"}
                    if chunk_res.copilot_stream_chunk:
                        chunk_callback(chunk_res.copilot_stream_chunk)
                    elif chunk_res.copilot_response_json:
                        return json.loads(chunk_res.copilot_response_json)
                except queue.Empty:
                    await asyncio.sleep(0.05)
        # Wrap in async loop
        async def run_remote_copilot():
            while True:
                action, commands, custom_cmd = await interface.run_session(
                    raw_bytes=bytes(client_buffer_bytes),
                    node_info=node_info,
                    on_ai_call=on_ai_call_remote,
                    blocks=blocks
                )
                if action == "continue":
                    # Send continue signal to server to loop back for another question
                    request_queue.put(connpy_pb2.InteractRequest(copilot_action="continue"))
                    continue
                return action, commands, custom_cmd
        with copilot_terminal_mode():
            action, commands, custom_cmd = asyncio.run(run_remote_copilot())
        print("\033[2m Returning to session...\033[0m", flush=True)
        # Prepare final action for server
        action_sent = "cancel"
        if action == "send_all" and commands:
            # In remote mode, send the selected commands as a custom block
            # so the server executes exactly what the user picked (e.g., selection '1')
            action_sent = f"custom:{chr(10).join(commands)}"
        elif action == "custom" and custom_cmd:
            action_sent = f"custom:{chr(10).join(custom_cmd)}"
        request_queue.put(connpy_pb2.InteractRequest(copilot_action=action_sent))
        resume_generator()
        tty.setraw(sys.stdin.fileno())
    @handle_errors
    def connect_node(self, unique_id, sftp=False, debug=False, logger=None):
        import sys
        import select
        import tty
        import termios
        import queue
        import os
        import threading
        request_queue = queue.Queue()
        client_buffer_bytes = bytearray()
        pause_stdin = [False]
        wake_r, wake_w = os.pipe()
        def pause_generator():
            pause_stdin[0] = True
            os.write(wake_w, b'\x00')
        def resume_generator():
            pause_stdin[0] = False
        def request_generator():
            cols, rows = 80, 24
            try:
@@ -63,8 +150,25 @@ class NodeStub:
            )
            while True:
-                r, _, _ = select.select([sys.stdin.fileno()], [], [])
+                try:
-                if r:
+                    while True:
                        req = request_queue.get_nowait()
                        if req is None:
                            return
                        yield req
                except queue.Empty:
                    pass
                if pause_stdin[0]:
                    import time
                    time.sleep(0.05)
                    continue
                r, _, _ = select.select([sys.stdin.fileno(), wake_r], [], [], 0.05)
                if wake_r in r:
                    os.read(wake_r, 1)
                    continue
                if sys.stdin.fileno() in r and not pause_stdin[0]:
                    try:
                        data = os.read(sys.stdin.fileno(), 1024)
                        if not data:
@@ -86,30 +190,74 @@ class NodeStub:
        old_tty = termios.tcgetattr(sys.stdin)
        try:
            import time
            tty.setraw(sys.stdin.fileno())
            response_iterator = self.stub.interact_node(request_generator())
-            # First response is connection status
+            import queue
            response_queue = queue.Queue()
            def response_consumer():
                try:
-                first_res = next(response_iterator)
+                    for r in response_iterator:
-                if first_res.success:
+                        response_queue.put(r)
                except Exception:
                    pass
                response_queue.put(None)
            t_consumer = threading.Thread(target=response_consumer, daemon=True)
            t_consumer.start()
            # First phase: Wait for connection status, print early data
            try:
                while True:
                    res = response_queue.get()
                    if res is None:
                        return
                    if res.stdout_data:
                        data = res.stdout_data
                        if debug:
                            data = data.replace(b'\x1b[H\x1b[2J', b'').replace(b'\x1bc', b'').replace(b'\x1b[3J', b'')
                        os.write(sys.stdout.fileno(), data)
                    if res.success:
                        # Connection established on server, show success message
                        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
                        printer.success(conn_msg)
                        pause_stdin[0] = False
                        tty.setraw(sys.stdin.fileno())
-                else:
+                        break
                    if res.error_message:
                        # Connection failed on server
                        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
-                    printer.error(f"Connection failed: {first_res.error_message}")
+                        printer.error(f"Connection failed: {res.error_message}")
                        return
-            except StopIteration:
+            except queue.Empty:
                return
-            for res in response_iterator:
+            # Second phase: Stream active session
            # Clear screen filter is only applied before success (Phase 1).
            # Once the user has a prompt, Ctrl+L must work normally.
            while True:
                res = response_queue.get()
                if res is None:
                    break
                if res.copilot_prompt:
                    self._handle_remote_copilot(
                        res, request_queue, response_queue, 
                        client_buffer_bytes, 
                        pause_generator, resume_generator, old_tty
                    )
                    continue
                if res.stdout_data:
                    os.write(sys.stdout.fileno(), res.stdout_data)
                    client_buffer_bytes.extend(res.stdout_data)
        finally:
            termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
            os.close(wake_r)
            os.close(wake_w)
    @handle_errors
    def connect_dynamic(self, connection_params, debug=False):
@@ -117,10 +265,22 @@ class NodeStub:
        import select
        import tty
        import termios
        import queue
        import os
        import json
        params_json = json.dumps(connection_params)
        request_queue = queue.Queue()
        client_buffer_bytes = bytearray()
        pause_stdin = [False]
        wake_r, wake_w = os.pipe()
        def pause_generator():
            pause_stdin[0] = True
            os.write(wake_w, b'\x00')
        def resume_generator():
            pause_stdin[0] = False
        def request_generator():
            cols, rows = 80, 24
@@ -136,8 +296,25 @@ class NodeStub:
            )
            while True:
-                r, _, _ = select.select([sys.stdin.fileno()], [], [])
+                try:
-                if r:
+                    while True:
                        req = request_queue.get_nowait()
                        if req is None:
                            return
                        yield req
                except queue.Empty:
                    pass
                if pause_stdin[0]:
                    import time
                    time.sleep(0.05)
                    continue
                r, _, _ = select.select([sys.stdin.fileno(), wake_r], [], [], 0.05)
                if wake_r in r:
                    os.read(wake_r, 1)
                    continue
                if sys.stdin.fileno() in r and not pause_stdin[0]:
                    try:
                        data = os.read(sys.stdin.fileno(), 1024)
                        if not data:
@@ -160,30 +337,72 @@ class NodeStub:
        old_tty = termios.tcgetattr(sys.stdin)
        try:
            import time
            tty.setraw(sys.stdin.fileno())
            response_iterator = self.stub.interact_node(request_generator())
-            # First response is connection status
+            import queue
            response_queue = queue.Queue()
            def response_consumer():
                try:
-                first_res = next(response_iterator)
+                    for r in response_iterator:
-                if first_res.success:
+                        response_queue.put(r)
                except Exception:
                    pass
                response_queue.put(None)
            t_consumer = threading.Thread(target=response_consumer, daemon=True)
            t_consumer.start()
            # First phase: Wait for connection status, print early data
            try:
                while True:
                    res = response_queue.get()
                    if res is None:
                        return
                    if res.stdout_data:
                        data = res.stdout_data
                        if debug:
                            data = data.replace(b'\x1b[H\x1b[2J', b'').replace(b'\x1bc', b'').replace(b'\x1b[3J', b'')
                        os.write(sys.stdout.fileno(), data)
                    if res.success:
                        # Connection established on server, show success message
                        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
                        printer.success(conn_msg)
                        pause_stdin[0] = False
                        tty.setraw(sys.stdin.fileno())
-                else:
+                        break
                    if res.error_message:
                        # Connection failed on server
                        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
-                    printer.error(f"Connection failed: {first_res.error_message}")
+                        printer.error(f"Connection failed: {res.error_message}")
                        return
-            except StopIteration:
+            except queue.Empty:
                return
-            for res in response_iterator:
+            # Second phase: Stream active session
            while True:
                res = response_queue.get()
                if res is None:
                    break
                if res.copilot_prompt:
                    self._handle_remote_copilot(
                        res, request_queue, response_queue, 
                        client_buffer_bytes, 
                        pause_generator, resume_generator, old_tty
                    )
                    continue
                if res.stdout_data:
                    os.write(sys.stdout.fileno(), res.stdout_data)
                    client_buffer_bytes.extend(res.stdout_data)
        finally:
            termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
            os.close(wake_r)
            os.close(wake_w)
    @MethodHook
    @handle_errors
@@ -508,7 +727,6 @@ class AIStub:
        import queue
        from rich.prompt import Prompt
        from rich.text import Text
        from rich.live import Live
        from rich.panel import Panel
        from rich.markdown import Markdown
@@ -539,7 +757,7 @@ class AIStub:
        responses = self.stub.ask(request_generator())
        full_content = ""
-        live_display = None
+        header_printed = False
        final_result = {"response": "", "chat_history": []}
        # Background thread to pull responses from gRPC into a local queue
@@ -604,69 +822,53 @@ class AIStub:
                if response.debug_message:
                    if debug:
                        if live_display:
                            try: live_display.stop()
                            except: pass
                        if status:
                            try: status.stop()
                            except: pass
                        printer.console.print(Text.from_ansi(response.debug_message))
-                        if live_display:
+                        if status:
                            try: live_display.start()
                            except: pass
                        elif status:
                            try: status.start()
                            except: pass
                    continue
                if response.important_message:
                    if live_display:
                        try: live_display.stop()
                        except: pass
                    if status:
                        try: status.stop()
                        except: pass
                    printer.console.print(Text.from_ansi(response.important_message))
-                    if live_display:
+                    if status:
                        try: live_display.start()
                        except: pass
                    elif status:
                        try: status.start()
                        except: pass
                    continue
                if not response.is_final:
                    if response.text_chunk:
-                        full_content += response.text_chunk
+                        if not header_printed:
                        if not live_display:
                            if status:
                                try: status.stop()
                                except: pass
                            from rich.console import Console as RichConsole
-                            from ..printer import connpy_theme, get_original_stdout
+                            from rich.rule import Rule
                            from ..printer import connpy_theme, get_original_stdout, IncrementalMarkdownParser
                            stable_console = RichConsole(theme=connpy_theme, file=get_original_stdout())
-                            # We default to Engineer title during stream, final result will correct it if needed
+                            # Print header on first chunk
-                            live_display = Live(
+                            stable_console.print(Rule("[bold engineer]Network Engineer[/bold engineer]", style="engineer"))
-                                Panel(Markdown(full_content), title="[bold engineer]Network Engineer[/bold engineer]", border_style="engineer", expand=False),
+                            header_printed = True
-                                console=stable_console,
+                            
-                                refresh_per_second=8,
+                            # Initialize parser
-                                transient=False
+                            md_parser = IncrementalMarkdownParser(console=stable_console)
-                            )
+                        
-                            live_display.start()
+                        full_content += response.text_chunk
-                        else:
+                        md_parser.feed(response.text_chunk)
                            live_display.update(
                                Panel(Markdown(full_content), title="[bold engineer]Network Engineer[/bold engineer]", border_style="engineer", expand=False)
                            )
                    continue
                if response.is_final:
-                    if live_display:
+                    if header_printed:
-                        try: live_display.stop()
+                        from rich.rule import Rule
-                        except: pass
+                        md_parser.flush()
-                    # Final stop for status to ensure it disappears before the panel
+                        
                    if status:
                        try: status.stop()
                        except: pass
@@ -677,13 +879,14 @@ class AIStub:
                    role_label = "Network Architect" if responder == "architect" else "Network Engineer"
                    title = f"[bold {alias}]{role_label}[/bold {alias}]"
-                    content_to_print = full_content or final_result.get("response", "")
+                    if header_printed:
-                    if content_to_print:
+                        from rich.console import Console as RichConsole
-                        if live_display:
+                        from ..printer import connpy_theme, get_original_stdout
-                            # Re-render the final frame with correct title/colors
+                        stable_console = RichConsole(theme=connpy_theme, file=get_original_stdout())
-                            live_display.update(Panel(Markdown(content_to_print), title=title, border_style=alias, expand=False))
+                        stable_console.print(Rule(style=alias))
-                        else:
+                    elif not full_content and final_result.get("response"):
-                            printer.console.print(Panel(Markdown(content_to_print), title=title, border_style=alias, expand=False))
+                        # If nothing streamed but we have response (e.g. error or direct guide)
                        printer.console.print(Panel(Markdown(final_result["response"]), title=title, border_style=alias, expand=False))
                    break
        except Exception as e:
            # Check if it was a gRPC error that we should let handle_errors catch
@@ -715,6 +918,17 @@ class AIStub:
        req = connpy_pb2.ProviderRequest(provider=provider, model=model or "", api_key=api_key or "")
        self.stub.configure_provider(req)
    @handle_errors
    def configure_mcp(self, name, url=None, enabled=True, auto_load_on_os=None, remove=False):
        req = connpy_pb2.MCPRequest(
            name=name,
            url=url or "",
            enabled=enabled,
            auto_load_on_os=auto_load_on_os or "",
            remove=remove
        )
        self.stub.configure_mcp(req)
    @handle_errors
    def load_session_data(self, session_id):
        return from_struct(self.stub.load_session_data(connpy_pb2.StringRequest(value=session_id)).data)
@@ -0,0 +1,171 @@
 import asyncio
 import json
 import os
 import threading
 from typing import Any, Dict, List, Optional
 import logging
 try:
    from mcp import ClientSession
    from mcp.client.sse import sse_client
    MCP_AVAILABLE = True
 except ImportError:
    MCP_AVAILABLE = False
 # Silence noisy MCP and HTTP internal logging
 logging.getLogger("mcp").setLevel(logging.CRITICAL)
 logging.getLogger("httpx").setLevel(logging.CRITICAL)
 logging.getLogger("httpcore").setLevel(logging.CRITICAL)
 class MCPClientManager:
    """Manages MCP SSE client connections for connpy."""
    _instance = None
    _lock = threading.Lock()
    def __new__(cls, *args, **kwargs):
        with cls._lock:
            if cls._instance is None:
                cls._instance = super(MCPClientManager, cls).__new__(cls)
                cls._instance._initialized = False
            return cls._instance
    def __init__(self, config=None):
        if self._initialized:
            return
        self.config = config
        self.sessions: Dict[str, Dict[str, Any]] = {} # name -> {session, stack}
        self.tool_cache: Dict[str, List[Dict[str, Any]]] = {}
        self._connecting: Dict[str, asyncio.Future] = {}
        self._initialized = True
    async def get_tools_for_llm(self, os_filter: Optional[str] = None) -> List[Dict[str, Any]]:
        """
        Fetches tools from enabled MCP servers that match the OS filter.
        """
        if not MCP_AVAILABLE:
            return []
        all_llm_tools = []
        try:
            mcp_config = self.config.config.get("ai", {}).get("mcp_servers", {})
        except Exception:
            return []
        async def _fetch(name, cfg):
            if not cfg.get("enabled", True): return []
            # Filter by OS if specified in config (primarily used for copilot strict matching)
            auto_os = cfg.get("auto_load_on_os")
            if os_filter is not None and auto_os and os_filter.lower() != auto_os.lower():
                return []
            try:
                session = await self._ensure_connected(name, cfg)
                if session:
                    if name in self.tool_cache: return self.tool_cache[name]
                    llm_tools = await self._fetch_tools_as_openai(name, session)
                    self.tool_cache[name] = llm_tools
                    return llm_tools
            except Exception:
                pass
            return []
        tasks = [ _fetch(name, cfg) for name, cfg in mcp_config.items() ]
        if tasks:
            results = await asyncio.gather(*tasks)
            for tools in results:
                all_llm_tools.extend(tools)
        return all_llm_tools
    async def _ensure_connected(self, name: str, cfg: Dict[str, Any]) -> Optional[Any]:
        if not MCP_AVAILABLE: return None
        if name in self.sessions and self.sessions[name].get("session"):
            return self.sessions[name]["session"]
        url = cfg.get("url")
        if not url:
            return None
        if name in self._connecting:
            try:
                return await asyncio.wait_for(asyncio.shield(self._connecting[name]), timeout=10.0)
            except Exception:
                return None
        loop = asyncio.get_running_loop()
        fut = loop.create_future()
        self._connecting[name] = fut
        try:
            from contextlib import AsyncExitStack
            stack = AsyncExitStack()
            async def _do_connect():
                read, write = await stack.enter_async_context(sse_client(url))
                session = await stack.enter_async_context(ClientSession(read, write))
                await session.initialize()
                return session
            session = await asyncio.wait_for(_do_connect(), timeout=15.0)
            self.sessions[name] = {"session": session, "stack": stack}
            fut.set_result(session)
            return session
        except Exception:
            fut.set_result(None)
            return None
        finally:
            if name in self._connecting:
                del self._connecting[name]
    async def _fetch_tools_as_openai(self, server_name: str, session: Any) -> List[Dict[str, Any]]:
        try:
            result = await asyncio.wait_for(session.list_tools(), timeout=5.0)
            openai_tools = []
            for tool in result.tools:
                # Use mcp_ prefix to ensure valid function name for LiteLLM/Gemini
                prefixed_name = f"mcp_{server_name}__{tool.name}"
                openai_tools.append({
                    "type": "function",
                    "function": {
                        "name": prefixed_name,
                        "description": f"[{server_name}] {tool.description}",
                        "parameters": tool.inputSchema
                    }
                })
            return openai_tools
        except Exception:
            return []
    async def call_tool(self, full_tool_name: str, arguments: Dict[str, Any]) -> Any:
        """Calls an MCP tool and returns text result."""
        if not MCP_AVAILABLE:
            return "Error: MCP SDK is not installed."
        if "__" not in full_tool_name:
            return f"Error: Tool {full_tool_name} is not a valid MCP tool."
        clean_name = full_tool_name[4:] if full_tool_name.startswith("mcp_") else full_tool_name
        server_name, tool_name = clean_name.split("__", 1)
        if server_name not in self.sessions:
            return f"Error: MCP server {server_name} is not connected."
        session = self.sessions[server_name]["session"]
        try:
            result = await asyncio.wait_for(session.call_tool(tool_name, arguments), timeout=60.0)
            text_outputs = [content.text for content in result.content if hasattr(content, "text")]
            return "\n".join(text_outputs) if text_outputs else str(result)
        except Exception as e:
            return f"Error calling tool {tool_name} on {server_name}: {str(e)}"
    async def shutdown(self):
        """Close all SSE connections."""
        for name, data in self.sessions.items():
            stack = data.get("stack")
            if stack:
                await stack.aclose()
        self.sessions = {}
@@ -15,7 +15,17 @@ class ThreadLocalStream:
    def write(self, data):
        stream = self._get_stream()
        if stream:
            import time
            retries = 0
            while True:
                try:
                    stream.write(data)
                    break
                except BlockingIOError:
                    if retries > 50:
                        raise
                    time.sleep(0.01)
                    retries += 1
    def flush(self):
        stream = self._get_stream()
@@ -46,8 +56,9 @@ def _get_local():
        _local.console = None
    if not hasattr(_local, 'err_console'):
        _local.err_console = None
-    if not hasattr(_local, 'theme'):
+    if not hasattr(_local, 'theme') or _local.theme is None:
-        _local.theme = None
+        from rich.theme import Theme
        _local.theme = Theme(_global_active_styles)
    return _local
 def set_thread_stream(stream):
@@ -69,23 +80,45 @@ def get_original_stderr():
 # Centralized design system
 STYLES = {
-    "info": "cyan",
+    "info": "#00ffff",        # Cyan
-    "warning": "yellow",
+    "warning": "#ffff00",     # Yellow
-    "error": "red",
+    "error": "#ff0000",       # Red
-    "success": "green",
+    "success": "#00ff00",     # Green
-    "debug": "dim",
+    "debug": "#888888",
-    "header": "bold cyan",
+    "header": "bold #00ffff",
-    "key": "bold cyan",
+    "key": "bold #00ffff",
-    "border": "cyan",
+    "border": "#00ffff",
-    "pass": "bold green",
+    "pass": "bold #00ff00",
-    "fail": "bold red",
+    "fail": "bold #ff0000",
-    "engineer": "blue",
+    "engineer": "#5fafff",    # Sky Blue (lighter than pure blue)
-    "architect": "medium_purple",
+    "architect": "#9370db",   # Medium Purple
-    "ai_status": "bold green",
+    "ai_status": "bold #00ff00",
-    "user_prompt": "bold cyan",
+    "user_prompt": "bold #00afd7", # Deep Sky Blue / Soft Cyan
-    "unavailable": "orange3",
+    "unavailable": "#d78700",
    "contrast": "#bbbbbb",
 }
 LIGHT_THEME = {
    "info": "#00008b",        # Navy Blue
    "warning": "#d78700",     # Orange
    "error": "#cd0000",       # Dark Red
    "success": "#006400",     # Dark Green
    "debug": "#777777",
    "header": "bold #00008b",
    "key": "bold #00008b",
    "border": "#00008b",
    "pass": "bold #006400",
    "fail": "bold #cd0000",
    "engineer": "#00008b",
    "architect": "#8b008b",   # Dark Magenta
    "ai_status": "bold #006400",
    "user_prompt": "bold #00008b",
    "unavailable": "#666666",
    "contrast": "#777777",
 }
 _global_active_styles = STYLES.copy()
 def _get_console():
    local = _get_local()
@@ -171,7 +204,7 @@ def connpy_theme():
    local = _get_local()
    if local.theme is None:
        from rich.theme import Theme
-        local.theme = Theme(STYLES)
+        local.theme = Theme(_global_active_styles)
    return local.theme
 def apply_theme(user_styles=None):
@@ -179,6 +212,7 @@ def apply_theme(user_styles=None):
    Updates the global console themes with user-defined styles.
    If a style is missing in user_styles, it falls back to the default in STYLES.
    """
    global _global_active_styles
    local = _get_local()
    from rich.theme import Theme
@@ -190,6 +224,7 @@ def apply_theme(user_styles=None):
            if key in active_styles:
                active_styles[key] = value
    _global_active_styles = active_styles
    local.theme = Theme(active_styles)
    if local.console:
        local.console.push_theme(local.theme)
@@ -202,10 +237,15 @@ def _format_multiline(tag, message, style=None):
    message = str(message)
    lines = message.splitlines()
    if not lines:
-        return f"[{style}]\\[{tag}][/{style}]" if style else f"\\[{tag}]"
+        if style:
            return f"[{style}]\\[{tag}][/{style}]"
        return f"\\[{tag}]"
    # Apply style to the tag if provided
    styled_tag = f"[{style}]\\[{tag}][/{style}]" if style else f"\\[{tag}]"
    if style:
        # Include brackets in the styling
        styled_tag = f"[{style}]\\[{tag}][/{style}]"
    formatted = [f"{styled_tag} {lines[0]}"]
    # Indent subsequent lines
@@ -462,7 +502,78 @@ class _ThemeProxy:
        local = _get_local()
        if local.theme is None:
            from rich.theme import Theme
-            local.theme = Theme(STYLES)
+            local.theme = Theme(_global_active_styles)
        return getattr(local.theme, name)
 connpy_theme = _ThemeProxy()
 class BlockMarkdownRenderer:
    """
    Block-buffered streaming markdown renderer.
    Accumulates text until block boundaries are detected,
    then renders complete blocks using Rich's Markdown.
    """
    def __init__(self, console=None):
        from rich.console import Console as RichConsole
        from .printer import connpy_theme, get_original_stdout
        self._console = console or RichConsole(
            theme=connpy_theme, file=get_original_stdout()
        )
        self._line_buf = ""        # chars waiting for \n
        self._block_lines = []     # complete lines for current block
        self._in_code_block = False
    def feed(self, text):
        self._line_buf += text
        while '\n' in self._line_buf:
            idx = self._line_buf.index('\n')
            line = self._line_buf[:idx + 1]
            self._line_buf = self._line_buf[idx + 1:]
            self._process_line(line)
    def flush(self):
        if self._line_buf:
            self._block_lines.append(self._line_buf)
            self._line_buf = ""
        self._flush_block()
    def _process_line(self, line):
        stripped = line.strip()
        if stripped.startswith('```'):
            if not self._in_code_block:
                # Flush accumulated text before code block
                self._flush_block()
                self._in_code_block = True
                self._block_lines.append(line)
            else:
                # Include closing fence and flush code block
                self._block_lines.append(line)
                self._in_code_block = False
                self._flush_block()
            return
        if self._in_code_block:
            self._block_lines.append(line)
            return
        # Blank line = paragraph break
        if stripped == '':
            self._block_lines.append(line)
            self._flush_block()
            return
        self._block_lines.append(line)
    def _flush_block(self):
        if not self._block_lines:
            return
        block_text = ''.join(self._block_lines).strip()
        self._block_lines = []
        if not block_text:
            return
        from rich.markdown import Markdown
        self._console.print(Markdown(block_text))
 # Alias for backward compatibility
 IncrementalMarkdownParser = BlockMarkdownRenderer
@@ -65,9 +65,11 @@ service ImportExportService {
 service AIService {
    rpc ask (stream AskRequest) returns (stream AIResponse) {}
    rpc confirm (StringRequest) returns (BoolResponse) {}
    rpc ask_copilot (CopilotRequest) returns (CopilotResponse) {}
    rpc list_sessions (google.protobuf.Empty) returns (ValueResponse) {}
    rpc delete_session (StringRequest) returns (google.protobuf.Empty) {}
    rpc configure_provider (ProviderRequest) returns (google.protobuf.Empty) {}
    rpc configure_mcp (MCPRequest) returns (google.protobuf.Empty) {}
    rpc load_session_data (StringRequest) returns (StructResponse) {}
 }
@@ -89,12 +91,24 @@ message InteractRequest {
    int32 cols = 5;
    int32 rows = 6;
    string connection_params_json = 7;
    // Copilot fields
    string copilot_question = 8;
    string copilot_action = 9;
    string copilot_context_buffer = 10;
    string copilot_node_info_json = 13;
 }
 message InteractResponse {
    bytes stdout_data = 1;
    bool success = 2;
    string error_message = 3;
    // Copilot fields
    bool copilot_prompt = 4;
    string copilot_buffer_preview = 5;
    string copilot_response_json = 6;
    string copilot_node_info_json = 7;
    string copilot_stream_chunk = 8;
    string copilot_injected_command = 9;
 }
 message FilterRequest {
@@ -257,3 +271,24 @@ message FullReplaceRequest {
    google.protobuf.Struct connections = 1;
    google.protobuf.Struct profiles = 2;
 }
 message CopilotRequest {
    string terminal_buffer = 1;
    string user_question = 2;
    string node_info_json = 3;
 }
 message CopilotResponse {
    repeated string commands = 1;
    string guide = 2;
    string risk_level = 3;
    string error = 4;
 }
 message MCPRequest {
    string name = 1;
    string url = 2;
    bool enabled = 3;
    string auto_load_on_os = 4;
    bool remove = 5;
 }
@@ -1,9 +1,143 @@
 import re
 from .base import BaseService
 from .exceptions import InvalidConfigurationError
 from connpy.utils import log_cleaner
 class AIService(BaseService):
    """Business logic for interacting with AI agents and LLM configurations."""
    def build_context_blocks(self, raw_bytes: bytes, cmd_byte_positions: list, node_info: dict, last_line: str = "") -> list:
        """Identifies command blocks in the terminal history."""
        blocks = []
        if not raw_bytes:
            return blocks
        default_prompt = r'>$|#$|\$$|>.$|#.$|\$.$'
        device_prompt = node_info.get("prompt", default_prompt) if isinstance(node_info, dict) else default_prompt
        prompt_re_str = re.sub(r'(?<!\\)\$', '', device_prompt)
        try:
            prompt_re = re.compile(prompt_re_str)
        except Exception:
            prompt_re = re.compile(re.sub(r'(?<!\\)\$', '', default_prompt))
        parsed_positions = []
        if cmd_byte_positions and len(cmd_byte_positions) >= 1:
            for i in range(1, len(cmd_byte_positions)):
                pos, known_cmd = cmd_byte_positions[i]
                prev_pos = cmd_byte_positions[i-1][0]
                if known_cmd:
                    prev_chunk = raw_bytes[prev_pos:pos]
                    prev_cleaned = log_cleaner(prev_chunk.decode(errors='replace'))
                    prev_lines = [l for l in prev_cleaned.split('\n') if l.strip()]
                    prompt_text = prev_lines[-1].strip() if prev_lines else ""
                    preview = f"{prompt_text}{known_cmd}" if prompt_text else known_cmd
                    parsed_positions.append({"pos": pos, "type": "VALID_CMD", "preview": preview[:80]})
                else:
                    chunk = raw_bytes[prev_pos:pos]
                    cleaned = log_cleaner(chunk.decode(errors='replace'))
                    lines = [l for l in cleaned.split('\n') if l.strip()]
                    preview = lines[-1].strip() if lines else ""
                    if preview:
                        match = prompt_re.search(preview)
                        if match:
                            cmd_text = preview[match.end():].strip()
                            if cmd_text:
                                parsed_positions.append({"pos": pos, "type": "VALID_CMD", "preview": preview[:80]})
                            else:
                                parsed_positions.append({"pos": pos, "type": "EMPTY_PROMPT", "preview": ""})
                        else:
                            parsed_positions.append({"pos": pos, "type": "SCROLLING", "preview": ""})
                    else:
                        parsed_positions.append({"pos": pos, "type": "SCROLLING", "preview": ""})
        last_newline = raw_bytes.rfind(b'\n')
        current_prompt_pos = last_newline + 1 if last_newline != -1 else 0
        current_end = len(raw_bytes)
        for i, item in enumerate(parsed_positions):
            if item["type"] == "VALID_CMD":
                start_pos = item["pos"]
                preview = item["preview"]
                # Find the end position: next VALID_CMD or EMPTY_PROMPT
                end_pos = current_prompt_pos
                for j in range(i + 1, len(parsed_positions)):
                    next_item = parsed_positions[j]
                    if next_item["type"] in ("VALID_CMD", "EMPTY_PROMPT"):
                        end_pos = next_item["pos"]
                        break
                blocks.append((start_pos, end_pos, preview))
        # Always ensure there is a final block representing the current prompt
        if not blocks:
            blocks.append((current_prompt_pos, current_end, last_line[:80] if last_line else "CURRENT CONTEXT"))
        elif blocks[-1][0] < current_prompt_pos:
            blocks.append((current_prompt_pos, current_end, last_line[:80] if last_line else "CURRENT CONTEXT"))
        return blocks
    def process_copilot_input(self, input_text: str, session_state: dict) -> dict:
        """Parses slash commands and manages session state. Returns directive dict."""
        text = input_text.strip()
        if not text.startswith('/'):
            return {"action": "execute", "clean_prompt": text, "overrides": {}}
        parts = text.split(maxsplit=1)
        cmd = parts[0].lower()
        args = parts[1] if len(parts) > 1 else ""
        # 1. State Commands (Persistent)
        if cmd == "/os":
            if args:
                session_state['os'] = args
                return {"action": "state_update", "message": f"OS context changed to {args}"}
        elif cmd == "/prompt":
            if args:
                session_state['prompt'] = args
                return {"action": "state_update", "message": f"Prompt regex changed to {args}"}
        elif cmd == "/memorize":
            if args:
                session_state['memories'].append(args)
                return {"action": "state_update", "message": f"Memory added: {args}"}
        elif cmd == "/clear":
            session_state['memories'] = []
            return {"action": "state_update", "message": "Memory cleared"}
        # 2. Hybrid Commands
        elif cmd == "/architect":
            if not args:
                session_state['persona'] = 'architect'
                return {"action": "state_update", "message": "Persona set to Architect"}
            else:
                return {"action": "execute", "clean_prompt": args, "overrides": {"persona": "architect"}}
        elif cmd == "/engineer":
            if not args:
                session_state['persona'] = 'engineer'
                return {"action": "state_update", "message": "Persona set to Engineer"}
            else:
                return {"action": "execute", "clean_prompt": args, "overrides": {"persona": "engineer"}}
        elif cmd == "/trust":
            if not args:
                session_state['trust_mode'] = True
                return {"action": "state_update", "message": "Auto-execute (trust) enabled for session"}
            else:
                return {"action": "execute", "clean_prompt": args, "overrides": {"trust": True}}
        elif cmd == "/untrust":
            if not args:
                session_state['trust_mode'] = False
                return {"action": "state_update", "message": "Auto-execute (trust) disabled for session"}
            else:
                return {"action": "execute", "clean_prompt": args, "overrides": {"trust": False}}
        # Unknown command, execute normally
        return {"action": "execute", "clean_prompt": text, "overrides": {}}
    def ask(self, input_text, dryrun=False, chat_history=None, status=None, debug=False, session_id=None, console=None, chunk_callback=None, confirm_handler=None, trust=False, **overrides):
        """Send a prompt to the AI agent."""
        from connpy.ai import ai
@@ -17,6 +151,21 @@ class AIService(BaseService):
        agent = ai(self.config, console=console)
        return agent.confirm(input_text)
    def ask_copilot(self, terminal_buffer, user_question, node_info=None, chunk_callback=None):
        """Ask the AI copilot for terminal assistance."""
        from connpy.ai import ai, run_ai_async
        agent = ai(self.config)
        future = run_ai_async(agent.aask_copilot(terminal_buffer, user_question, node_info, chunk_callback=chunk_callback))
        return future.result()
    async def aask_copilot(self, terminal_buffer, user_question, node_info=None, chunk_callback=None):
        """Ask the AI copilot for terminal assistance asynchronously."""
        from connpy.ai import ai, run_ai_async
        import asyncio
        agent = ai(self.config)
        future = run_ai_async(agent.aask_copilot(terminal_buffer, user_question, node_info, chunk_callback=chunk_callback))
        return await asyncio.wrap_future(future)
    def list_sessions(self):
        """Return a list of all saved AI sessions."""
@@ -45,6 +194,40 @@ class AIService(BaseService):
        self.config.config["ai"] = settings
        self.config._saveconfig(self.config.file)
    def configure_mcp(self, name, url=None, enabled=None, auto_load_on_os=None, remove=False):
        """Update MCP server settings in the configuration with smart merging."""
        ai_settings = self.config.config.get("ai", {})
        mcp_servers = ai_settings.get("mcp_servers", {})
        if remove:
            if name in mcp_servers:
                del mcp_servers[name]
        else:
            # Get existing or new
            server_cfg = mcp_servers.get(name, {})
            # Partial updates
            if url is not None:
                server_cfg["url"] = url
            if enabled is not None:
                server_cfg["enabled"] = bool(enabled)
            elif "enabled" not in server_cfg:
                server_cfg["enabled"] = True # Default for new entries
            if auto_load_on_os is not None:
                if auto_load_on_os == "": # Explicit clear
                    if "auto_load_on_os" in server_cfg:
                        del server_cfg["auto_load_on_os"]
                else:
                    server_cfg["auto_load_on_os"] = auto_load_on_os
            mcp_servers[name] = server_cfg
        ai_settings["mcp_servers"] = mcp_servers
        self.config.config["ai"] = ai_settings
        self.config._saveconfig(self.config.file)
    def load_session_data(self, session_id):
        """Load a session's raw data by ID."""
        from connpy.ai import ai
@@ -70,6 +70,10 @@ class ConfigService(BaseService):
        if not isinstance(user_styles, dict):
            raise InvalidConfigurationError("Theme file must be a YAML dictionary.")
        # Support both direct styles and nested under 'theme' key
        if "theme" in user_styles and isinstance(user_styles["theme"], dict):
            user_styles = user_styles["theme"]
        # Filter for valid styles only (prevent junk in config)
        valid_styles = {k: v for k, v in user_styles.items() if k in STYLES}
@@ -14,7 +14,7 @@ class ExecutionService(BaseService):
        commands: List[str], 
        variables: Optional[Dict[str, Any]] = None,
        parallel: int = 10,
-        timeout: int = 10,
+        timeout: int = 20,
        folder: Optional[str] = None,
        prompt: Optional[str] = None,
        on_node_complete: Optional[Callable] = None,
@@ -62,7 +62,7 @@ class ExecutionService(BaseService):
        expected: List[str],
        variables: Optional[Dict[str, Any]] = None,
        parallel: int = 10,
-        timeout: int = 10,
+        timeout: int = 20,
        folder: Optional[str] = None,
        prompt: Optional[str] = None,
        on_node_complete: Optional[Callable] = None,
@@ -139,7 +139,7 @@ class ExecutionService(BaseService):
            "commands": playbook["commands"],
            "variables": playbook.get("variables"),
            "parallel": options.get("parallel", parallel),
-            "timeout": playbook.get("timeout", options.get("timeout", 10)),
+            "timeout": playbook.get("timeout", options.get("timeout", 20)),
            "prompt": options.get("prompt"),
            "name": playbook.get("name", "Task")
        }
@@ -233,14 +233,26 @@ class PluginService(BaseService):
        from rich.console import Console
        from rich.console import Console
-        buf = io.StringIO()
+        import queue
        import threading
        q = queue.Queue()
        class QueueIO(io.StringIO):
            def write(self, s):
                q.put(s)
                return len(s)
            def flush(self):
                pass
        buf = QueueIO()
        old_console = printer._get_console()
        old_err_console = printer._get_err_console()
        def run_plugin():
            printer.set_thread_console(Console(file=buf, theme=printer.connpy_theme, force_terminal=True))
            printer.set_thread_err_console(Console(file=buf, theme=printer.connpy_theme, force_terminal=True))
            printer.set_thread_stream(buf)
            try:
                if hasattr(module, "Entrypoint"):
                    module.Entrypoint(args, parser, app)
@@ -252,6 +264,13 @@ class PluginService(BaseService):
                printer.set_thread_console(old_console)
                printer.set_thread_err_console(old_err_console)
                printer.set_thread_stream(None)
                q.put(None)
-        for line in buf.getvalue().splitlines(keepends=True):
+        t = threading.Thread(target=run_plugin, daemon=True)
-            yield line
+        t.start()
        while True:
            item = q.get()
            if item is None:
                break
            yield item
@@ -0,0 +1,160 @@
 import pytest
 from unittest.mock import MagicMock, patch, AsyncMock
 import json
 import asyncio
 from connpy.ai import ai
 from connpy.core import node
 class DummyConfig:
    def __init__(self):
        self.config = {"ai": {"engineer_api_key": "test_key", "engineer_model": "test_model"}}
        self.defaultdir = "/tmp"
 class MockAsyncIterator:
    def __init__(self, items):
        self.items = items
    def __aiter__(self):
        return self
    async def __anext__(self):
        if not self.items:
            raise StopAsyncIteration
        return self.items.pop(0)
@pytest.fixture
 def mock_acompletion():
    # Patch acompletion inside connpy.ai.aask_copilot
    with patch('litellm.acompletion') as mock:
        yield mock
 def test_aask_copilot_tool_call(mock_acompletion):
    agent = ai(DummyConfig())
    # Setup mock response for streaming
    class MockDelta:
        def __init__(self, content):
            self.content = content
    class MockChoice:
        def __init__(self, content):
            self.delta = MockDelta(content)
    class MockChunk:
        def __init__(self, content):
            self.choices = [MockChoice(content)]
    # acompletion is awaited and returns an async iterator
    async def mock_ac(*args, **kwargs):
        return MockAsyncIterator([
            MockChunk("<guide>Check the interfaces and running config.</guide>"),
            MockChunk("<commands>\nshow ip int br\nshow run\n</commands>"),
            MockChunk("<risk>low</risk>")
        ])
    mock_acompletion.side_effect = mock_ac
    async def run_test():
        return await agent.aask_copilot("Router#", "What do I do?")
    result = asyncio.run(run_test())
    if result["error"]:
        print(f"ERROR OCCURRED: {result['error']}")
    assert result["error"] is None
    assert result["guide"] == "Check the interfaces and running config."
    assert result["risk_level"] == "low"
    assert result["commands"] == ["show ip int br", "show run"]
 def test_aask_copilot_fallback(mock_acompletion):
    agent = ai(DummyConfig())
    # Setup mock response for streaming
    class MockDelta:
        def __init__(self, content):
            self.content = content
    class MockChoice:
        def __init__(self, content):
            self.delta = MockDelta(content)
    class MockChunk:
        def __init__(self, content):
            self.choices = [MockChoice(content)]
    async def mock_ac(*args, **kwargs):
        return MockAsyncIterator([
            MockChunk("Here is some text response instead of tool call.")
        ])
    mock_acompletion.side_effect = mock_ac
    async def run_test():
        return await agent.aask_copilot("Router#", "What do I do?")
    result = asyncio.run(run_test())
    if result["error"]:
        print(f"ERROR OCCURRED: {result['error']}")
    assert result["error"] is None
    assert result["guide"] == "Here is some text response instead of tool call."
    assert result["risk_level"] == "low"
 def test_logclean_ansi():
    c = node("test_node", "1.2.3.4")
    raw = "Router#\x1b[K\x1b[m show ip"
    clean = c._logclean(raw, var=True)
    assert "\x1b" not in clean
 def test_ingress_task_interception():
    async def run_test():
        c = node("test_node", "1.2.3.4")
        c.mylog = MagicMock()
        c.mylog.getvalue.return_value = b"Some session log"
        c.unique = "test_node"
        c.host = "1.2.3.4"
        c.tags = {"os": "cisco_ios"}
        class MockStream:
            def __init__(self):
                self.data = [b"a", b"b", b"\x00", b"c", b""]
            async def read(self):
                if self.data:
                    return self.data.pop(0)
                return b""
            def setup(self, resize_callback):
                pass
        stream = MockStream()
        called_copilot = False
        async def mock_handler(buffer, node_info, s, child_fd):
            nonlocal called_copilot
            called_copilot = True
            assert buffer == "Some session log"
            assert node_info["os"] == "cisco_ios"
        c.child = MagicMock()
        c.child.child_fd = 123
        c.child.after = b""
        c.child.buffer = b""
        async def mock_ingress():
            while True:
                data = await stream.read()
                if not data:
                    break
                if mock_handler and b'\x00' in data:
                    buffer = c.mylog.getvalue().decode()
                    node_info = {"name": getattr(c, 'unique', 'unknown'), "host": getattr(c, 'host', 'unknown')}
                    if isinstance(getattr(c, 'tags', None), dict):
                        node_info["os"] = c.tags.get("os", "unknown")
                    await mock_handler(buffer, node_info, stream, c.child.child_fd)
                    continue
        await mock_ingress()
        assert called_copilot
    asyncio.run(run_test())
@@ -85,8 +85,8 @@ class TestStubsMessageFormatting:
        mock_resp = MagicMock()
        mock_resp.success = True
        mock_resp.stdout_data = b''
        stub.stub.interact_node.return_value = iter([mock_resp])        
        with patch("connpy.printer.success") as mock_success:
            with patch("sys.stdin.fileno", return_value=0):
                mock_select.return_value = ([], [], [])
@@ -47,6 +47,24 @@ class LocalStream:
                # signal handling not supported on some loops (e.g., Windows Proactor)
                pass
    def stop_reading(self):
        """Temporarily stop reading from stdin."""
        if self._loop and self.stdin_fd is not None:
            try:
                self._loop.remove_reader(self.stdin_fd)
            except Exception:
                pass
    def start_reading(self):
        """Resume reading from stdin."""
        if self._loop and self.stdin_fd is not None:
            try:
                # Ensure we don't add it twice
                self._loop.remove_reader(self.stdin_fd)
            except Exception:
                pass
            self._loop.add_reader(self.stdin_fd, self._read_ready)
    def teardown(self):
        if self._loop:
            try:
@@ -127,6 +145,7 @@ class RemoteStream:
        self.response_queue = response_queue
        self.running = True
        self._reader_queue = asyncio.Queue()
        self.copilot_queue = asyncio.Queue()
        self.resize_callback = None
        self._loop = None
        self.t = None
@@ -143,6 +162,19 @@ class RemoteStream:
                    if req.cols > 0 and req.rows > 0:
                        if self.resize_callback:
                            self._loop.call_soon_threadsafe(self.resize_callback, req.rows, req.cols)
                    # Copilot dispatching
                    copilot_msg = {}
                    if getattr(req, "copilot_question", ""):
                        copilot_msg.update({
                            "question": req.copilot_question,
                            "context_buffer": getattr(req, "copilot_context_buffer", ""),
                            "node_info_json": getattr(req, "copilot_node_info_json", "")
                        })
                    if getattr(req, "copilot_action", ""):
                        copilot_msg["action"] = req.copilot_action
                    if copilot_msg:
                        self._loop.call_soon_threadsafe(self.copilot_queue.put_nowait, copilot_msg)
                    if req.stdin_data:
                        self._loop.call_soon_threadsafe(self._reader_queue.put_nowait, req.stdin_data)
            except Exception:
@@ -0,0 +1,67 @@
 import re
 def log_cleaner(data: str) -> str:
    """
    Stateless utility to remove ANSI sequences and process cursor movements.
    """
    if not data:
        return ""
    lines = data.split('\n')
    cleaned_lines = []
    # Regex to capture: ANSI sequences, control characters (\r, \b, etc), and plain text chunks
    token_re = re.compile(r'(\x1B(?:[@-Z\\-_]|\[[0-?]*[ -/ ]*[@-~])|\r|\b|\x7f|[\x00-\x1F]|[^\x1B\r\b\x7f\x00-\x1F]+)')
    for line in lines:
        buffer = []
        cursor = 0
        for token in token_re.findall(line):
            if token == '\r':
                cursor = 0
            elif token in ('\b', '\x7f'):
                if cursor > 0:
                    cursor -= 1
            elif token.startswith('\x1B[') and len(token) >= 3:
                # Parse CSI: \x1B[ <params> <final_char>
                final = token[-1]
                param_str = token[2:-1]
                n = int(param_str) if param_str.isdigit() else 1
                if final == 'D':   # CUB – Cursor Back
                    cursor = max(0, cursor - n)
                elif final == 'C': # CUF – Cursor Forward
                    cursor = min(len(buffer), cursor + n)
                elif final == 'K': # EL  – Erase in Line
                    if n == 0 or param_str == '':      # Clear to end
                        buffer = buffer[:cursor]
                    elif n == 1:                         # Clear to start
                        buffer[:cursor] = [' '] * cursor
                    elif n == 2:                         # Clear entire line
                        buffer = []
                        cursor = 0
                elif final == 'G': # CHA – Cursor Horizontal Absolute (1-indexed)
                    cursor = max(0, n - 1)
                    # Pad buffer if cursor is beyond current length
                    if cursor > len(buffer):
                        buffer.extend([' '] * (cursor - len(buffer)))
                elif final == 'P': # DCH – Delete Characters
                    del buffer[cursor:cursor + n]
                elif final == '@': # ICH – Insert Characters
                    buffer[cursor:cursor] = [' '] * n
                # All other CSI sequences are silently discarded
            elif token.startswith('\x1B'):
                continue
            elif len(token) == 1 and ord(token) < 32:
                continue
            else:
                for char in token:
                    if cursor == len(buffer):
                        buffer.append(char)
                    else:
                        buffer[cursor] = char
                    cursor += 1
        cleaned_lines.append("".join(buffer))
    return "\n".join(cleaned_lines).replace('\n\n', '\n').strip()
@@ -1,9 +1,17 @@
 version: "3.8"
 services:
  connpy-app:
    build: .
-    image: connpy-app
+    image: connpy:latest
    container_name: connpy
    # Fundamental para la interactividad de la terminal
    stdin_open: true
    tty: true
    environment:
      - TERM=xterm-256color
    extra_hosts:
      - "host.docker.internal:host-gateway"
    volumes:
-      - ./docker/connpy/:/app
+      - ./docker/config:/config
-      - ./docker/logs/:/logs
+      - ./docker/ssh:/root/.ssh
-      - ./docker/ssh/:/root/.ssh/
+      - /var/run/docker.sock:/var/run/docker.sock
    # No definimos comando por defecto para que 'run' sea más natural
@@ -1,21 +1,65 @@
-# Use the official python image
+# connpy v6.0.0b8 - Modern Network Automation Environment (Local Build)
 FROM python:3.11-slim
-FROM python:3.11-alpine as connpy-app
+LABEL description="Connpy: AI-Driven Network Automation & Intelligence Platform"
 # Configuración de Terminal y Python
 ENV DEBIAN_FRONTEND=noninteractive \
    PYTHONUNBUFFERED=1 \
    TERM=xterm-256color
 # Set the entrypoint
 # Set the working directory
 WORKDIR /app
-# Install any additional dependencies
+# 1. Herramientas base del sistema
-RUN apk update && apk add --no-cache openssh fzf fzf-tmux ncurses bash
+RUN apt-get update && apt-get install -y --no-install-recommends \
-RUN pip3 install connpy
+    curl \
-RUN connpy config --configfolder /app
+    git \
    openssh-client \
    fzf \
    ncurses-bin \
    bash \
    procps \
    unzip \
    ca-certificates \
    gnupg \
    iputils-ping \
    telnet \
    && rm -rf /var/lib/apt/lists/*
-#AUTH
+# 2. Instalar Docker CLI (para el plugin de docker de connpy)
-RUN ssh-keygen -A
+RUN install -m 0755 -d /etc/apt/keyrings && \
-RUN mkdir /root/.ssh && \
+    curl -fsSL https://download.docker.com/linux/debian/gpg | gpg --dearmor -o /etc/apt/keyrings/docker.gpg && \
-    chmod 700 /root/.ssh
+    echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/debian $(. /etc/os-release && echo "$VERSION_CODENAME") stable" | \
    tee /etc/apt/sources.list.d/docker.list > /dev/null && \
    apt-get update && apt-get install -y docker-ce-cli && \
    rm -rf /var/lib/apt/lists/*
 # 3. Instalar Kubectl (para el plugin de k8s de connpy)
 RUN curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/$(dpkg --print-architecture)/kubectl" && \
    install -o root -g root -m 0755 kubectl /usr/local/bin/kubectl && \
    rm kubectl
-#Set the entrypoint
+# 4. Instalar AWS CLI y Session Manager Plugin (Universal x86_64/ARM64)
-ENTRYPOINT ["connpy"]
+RUN ARCH=$(uname -m) && \
    if [ "$ARCH" = "x86_64" ]; then AWS_ARCH="x86_64"; else AWS_ARCH="aarch64"; fi && \
    curl "https://awscli.amazonaws.com/awscli-exe-linux-$AWS_ARCH.zip" -o "awscliv2.zip" && \
    unzip awscliv2.zip && ./aws/install && rm -rf awscliv2.zip aws/ && \
    if [ "$ARCH" = "x86_64" ]; then \
        curl "https://s3.amazonaws.com/session-manager-downloads/plugin/latest/ubuntu_64bit/session-manager-plugin.deb" -o "ssm.deb"; \
    else \
        curl "https://s3.amazonaws.com/session-manager-downloads/plugin/latest/ubuntu_arm64/session-manager-plugin.deb" -o "ssm.deb"; \
    fi && \
    dpkg -i ssm.deb && rm ssm.deb
 # 5. Copiar código local e instalar dependencias
 COPY . .
 RUN pip install --no-cache-dir --upgrade pip && \
    pip install --no-cache-dir .
 # 6. Configuración de persistencia
 # Creamos la carpeta y el puntero .folder para que connpy use /config
 RUN mkdir -p /config /root/.ssh /root/.config/conn && chmod 700 /root/.ssh && \
    echo -n "/config" > /root/.config/conn/.folder
 # Punto de entrada directo a connpy
 ENTRYPOINT ["conn"]
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.ai_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -78,6 +78,9 @@ el.replaceWith(d);
                printer.error(str(e))
            return
        if args.mcp is not None:
            return self.configure_mcp(args)
        # Determinar session_id para retomar
        session_id = None
        if args.resume:
@@ -156,7 +159,7 @@ el.replaceWith(d);
            try:
                user_query = Prompt.ask(&#34;[user_prompt]User[/user_prompt]&#34;)
                if not user_query.strip(): continue
-                if user_query.lower() in [&#39;exit&#39;, &#39;quit&#39;, &#39;bye&#39;]: break
+                if user_query.lower() in [&#39;exit&#39;, &#39;quit&#39;, &#39;bye&#39;, &#39;cancel&#39;]: break
                with console.status(&#34;[ai_status]Agent is thinking...&#34;) as status:
                    result = self.app.myai.ask(user_query, chat_history=history, status=status, debug=args.debug, trust=args.trust, **self.ai_overrides)
@@ -179,11 +182,245 @@ el.replaceWith(d);
                    console.print(f&#34;[debug]Tokens: {u[&#39;total&#39;]} (Input: {u[&#39;input&#39;]}, Output: {u[&#39;output&#39;]})[/debug]&#34;)
            except (KeyboardInterrupt, EOFError):
                console.print(&#34;\n[dim]Session closed.[/dim]&#34;)
-                break</code></pre>
+                break
    def configure_mcp(self, args):
        &#34;&#34;&#34;Handle MCP server configuration via CLI tokens or interactive wizard.&#34;&#34;&#34;
        mcp_args = args.mcp
        # 1. Non-interactive CLI Mode (if arguments are provided)
        if mcp_args:
            action = mcp_args[0].lower()
            if action == &#34;list&#34;:
                settings = self.app.services.config_svc.get_settings()
                mcp_servers = settings.get(&#34;ai&#34;, {}).get(&#34;mcp_servers&#34;, {})
                if not mcp_servers:
                    printer.info(&#34;No MCP servers configured.&#34;)
                else:
                    columns = [&#34;Name&#34;, &#34;URL&#34;, &#34;Enabled&#34;, &#34;Auto-load OS&#34;]
                    rows = []
                    for name, cfg in mcp_servers.items():
                        rows.append([
                            name, 
                            cfg.get(&#34;url&#34;, &#34;&#34;), 
                            &#34;[green]Yes[/green]&#34; if cfg.get(&#34;enabled&#34;, True) else &#34;[red]No[/red]&#34;,
                            cfg.get(&#34;auto_load_on_os&#34;, &#34;Any&#34;)
                        ])
                    printer.table(&#34;Configured MCP Servers&#34;, columns, rows)
                return
            elif action == &#34;add&#34;:
                if len(mcp_args) &lt; 3:
                    printer.error(&#34;Usage: connpy ai --mcp add &lt;name&gt; &lt;url&gt; [os_filter]&#34;)
                    return
                name, url = mcp_args[1], mcp_args[2]
                os_filter = mcp_args[3] if len(mcp_args) &gt; 3 else None
                try:
                    self.app.services.ai.configure_mcp(name, url=url, auto_load_on_os=os_filter)
                    printer.success(f&#34;MCP server &#39;{name}&#39; added/updated.&#34;)
                except Exception as e:
                    printer.error(str(e))
                return
            elif action == &#34;remove&#34;:
                if len(mcp_args) &lt; 2:
                    printer.error(&#34;Usage: connpy ai --mcp remove &lt;name&gt;&#34;)
                    return
                name = mcp_args[1]
                try:
                    self.app.services.ai.configure_mcp(name, remove=True)
                    printer.success(f&#34;MCP server &#39;{name}&#39; removed.&#34;)
                except Exception as e:
                    printer.error(str(e))
                return
            elif action in [&#34;enable&#34;, &#34;disable&#34;]:
                if len(mcp_args) &lt; 2:
                    printer.error(f&#34;Usage: connpy ai --mcp {action} &lt;name&gt;&#34;)
                    return
                name = mcp_args[1]
                enabled = (action == &#34;enable&#34;)
                try:
                    self.app.services.ai.configure_mcp(name, enabled=enabled)
                    printer.success(f&#34;MCP server &#39;{name}&#39; {&#39;enabled&#39; if enabled else &#39;disabled&#39;}.&#34;)
                except Exception as e:
                    printer.error(str(e))
                return
            else:
                printer.error(f&#34;Unknown MCP action: {action}&#34;)
                printer.info(&#34;Available actions: list, add, remove, enable, disable&#34;)
                return
        # 2. Interactive Wizard Mode (if no arguments provided)
        # Import forms dynamically to avoid circular dependencies if any
        if not hasattr(self.app, &#34;cli_forms&#34;):
            from .forms import Forms
            self.app.cli_forms = Forms(self.app)
        settings = self.app.services.config_svc.get_settings()
        mcp_servers = settings.get(&#34;ai&#34;, {}).get(&#34;mcp_servers&#34;, {})
        result = self.app.cli_forms.mcp_wizard(mcp_servers)
        if not result:
            return
        action = result[&#34;action&#34;]
        try:
            if action == &#34;list&#34;:
                # Recursive call to the non-interactive list logic
                args.mcp = [&#34;list&#34;]
                return self.configure_mcp(args)
            elif action == &#34;add&#34;:
                self.app.services.ai.configure_mcp(
                    result[&#34;name&#34;], 
                    url=result[&#34;url&#34;], 
                    enabled=result[&#34;enabled&#34;],
                    auto_load_on_os=result[&#34;os&#34;]
                )
                printer.success(f&#34;MCP server &#39;{result[&#39;name&#39;]}&#39; saved.&#34;)
            elif action == &#34;update&#34;: # Used for toggle
                self.app.services.ai.configure_mcp(
                    result[&#34;name&#34;], 
                    enabled=result[&#34;enabled&#34;]
                )
                printer.success(f&#34;MCP server &#39;{result[&#39;name&#39;]}&#39; updated.&#34;)
            elif action == &#34;remove&#34;:
                self.app.services.ai.configure_mcp(result[&#34;name&#34;], remove=True)
                printer.success(f&#34;MCP server &#39;{result[&#39;name&#39;]}&#39; removed.&#34;)
        except Exception as e:
            printer.error(str(e))</code></pre>
 </details>
 <div class="desc"></div>
 <h3>Methods</h3>
 <dl>
 <dt id="connpy.cli.ai_handler.AIHandler.configure_mcp"><code class="name flex">
 <span>def <span class="ident">configure_mcp</span></span>(<span>self, args)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def configure_mcp(self, args):
    &#34;&#34;&#34;Handle MCP server configuration via CLI tokens or interactive wizard.&#34;&#34;&#34;
    mcp_args = args.mcp
    # 1. Non-interactive CLI Mode (if arguments are provided)
    if mcp_args:
        action = mcp_args[0].lower()
        if action == &#34;list&#34;:
            settings = self.app.services.config_svc.get_settings()
            mcp_servers = settings.get(&#34;ai&#34;, {}).get(&#34;mcp_servers&#34;, {})
            if not mcp_servers:
                printer.info(&#34;No MCP servers configured.&#34;)
            else:
                columns = [&#34;Name&#34;, &#34;URL&#34;, &#34;Enabled&#34;, &#34;Auto-load OS&#34;]
                rows = []
                for name, cfg in mcp_servers.items():
                    rows.append([
                        name, 
                        cfg.get(&#34;url&#34;, &#34;&#34;), 
                        &#34;[green]Yes[/green]&#34; if cfg.get(&#34;enabled&#34;, True) else &#34;[red]No[/red]&#34;,
                        cfg.get(&#34;auto_load_on_os&#34;, &#34;Any&#34;)
                    ])
                printer.table(&#34;Configured MCP Servers&#34;, columns, rows)
            return
        elif action == &#34;add&#34;:
            if len(mcp_args) &lt; 3:
                printer.error(&#34;Usage: connpy ai --mcp add &lt;name&gt; &lt;url&gt; [os_filter]&#34;)
                return
            name, url = mcp_args[1], mcp_args[2]
            os_filter = mcp_args[3] if len(mcp_args) &gt; 3 else None
            try:
                self.app.services.ai.configure_mcp(name, url=url, auto_load_on_os=os_filter)
                printer.success(f&#34;MCP server &#39;{name}&#39; added/updated.&#34;)
            except Exception as e:
                printer.error(str(e))
            return
        elif action == &#34;remove&#34;:
            if len(mcp_args) &lt; 2:
                printer.error(&#34;Usage: connpy ai --mcp remove &lt;name&gt;&#34;)
                return
            name = mcp_args[1]
            try:
                self.app.services.ai.configure_mcp(name, remove=True)
                printer.success(f&#34;MCP server &#39;{name}&#39; removed.&#34;)
            except Exception as e:
                printer.error(str(e))
            return
        elif action in [&#34;enable&#34;, &#34;disable&#34;]:
            if len(mcp_args) &lt; 2:
                printer.error(f&#34;Usage: connpy ai --mcp {action} &lt;name&gt;&#34;)
                return
            name = mcp_args[1]
            enabled = (action == &#34;enable&#34;)
            try:
                self.app.services.ai.configure_mcp(name, enabled=enabled)
                printer.success(f&#34;MCP server &#39;{name}&#39; {&#39;enabled&#39; if enabled else &#39;disabled&#39;}.&#34;)
            except Exception as e:
                printer.error(str(e))
            return
        else:
            printer.error(f&#34;Unknown MCP action: {action}&#34;)
            printer.info(&#34;Available actions: list, add, remove, enable, disable&#34;)
            return
    # 2. Interactive Wizard Mode (if no arguments provided)
    # Import forms dynamically to avoid circular dependencies if any
    if not hasattr(self.app, &#34;cli_forms&#34;):
        from .forms import Forms
        self.app.cli_forms = Forms(self.app)
    settings = self.app.services.config_svc.get_settings()
    mcp_servers = settings.get(&#34;ai&#34;, {}).get(&#34;mcp_servers&#34;, {})
    result = self.app.cli_forms.mcp_wizard(mcp_servers)
    if not result:
        return
    action = result[&#34;action&#34;]
    try:
        if action == &#34;list&#34;:
            # Recursive call to the non-interactive list logic
            args.mcp = [&#34;list&#34;]
            return self.configure_mcp(args)
        elif action == &#34;add&#34;:
            self.app.services.ai.configure_mcp(
                result[&#34;name&#34;], 
                url=result[&#34;url&#34;], 
                enabled=result[&#34;enabled&#34;],
                auto_load_on_os=result[&#34;os&#34;]
            )
            printer.success(f&#34;MCP server &#39;{result[&#39;name&#39;]}&#39; saved.&#34;)
        elif action == &#34;update&#34;: # Used for toggle
            self.app.services.ai.configure_mcp(
                result[&#34;name&#34;], 
                enabled=result[&#34;enabled&#34;]
            )
            printer.success(f&#34;MCP server &#39;{result[&#39;name&#39;]}&#39; updated.&#34;)
        elif action == &#34;remove&#34;:
            self.app.services.ai.configure_mcp(result[&#34;name&#34;], remove=True)
            printer.success(f&#34;MCP server &#39;{result[&#39;name&#39;]}&#39; removed.&#34;)
    except Exception as e:
        printer.error(str(e))</code></pre>
 </details>
 <div class="desc"><p>Handle MCP server configuration via CLI tokens or interactive wizard.</p></div>
 </dd>
 <dt id="connpy.cli.ai_handler.AIHandler.dispatch"><code class="name flex">
 <span>def <span class="ident">dispatch</span></span>(<span>self, args)</span>
 </code></dt>
@@ -211,6 +448,9 @@ el.replaceWith(d);
            printer.error(str(e))
        return
    if args.mcp is not None:
        return self.configure_mcp(args)
    # Determinar session_id para retomar
    session_id = None
    if args.resume:
@@ -283,7 +523,7 @@ el.replaceWith(d);
        try:
            user_query = Prompt.ask(&#34;[user_prompt]User[/user_prompt]&#34;)
            if not user_query.strip(): continue
-            if user_query.lower() in [&#39;exit&#39;, &#39;quit&#39;, &#39;bye&#39;]: break
+            if user_query.lower() in [&#39;exit&#39;, &#39;quit&#39;, &#39;bye&#39;, &#39;cancel&#39;]: break
            with console.status(&#34;[ai_status]Agent is thinking...&#34;) as status:
                result = self.app.myai.ask(user_query, chat_history=history, status=status, debug=args.debug, trust=args.trust, **self.ai_overrides)
@@ -356,6 +596,7 @@ el.replaceWith(d);
 <li>
 <h4><code><a title="connpy.cli.ai_handler.AIHandler" href="#connpy.cli.ai_handler.AIHandler">AIHandler</a></code></h4>
 <ul class="">
 <li><code><a title="connpy.cli.ai_handler.AIHandler.configure_mcp" href="#connpy.cli.ai_handler.AIHandler.configure_mcp">configure_mcp</a></code></li>
 <li><code><a title="connpy.cli.ai_handler.AIHandler.dispatch" href="#connpy.cli.ai_handler.AIHandler.dispatch">dispatch</a></code></li>
 <li><code><a title="connpy.cli.ai_handler.AIHandler.interactive_chat" href="#connpy.cli.ai_handler.AIHandler.interactive_chat">interactive_chat</a></code></li>
 <li><code><a title="connpy.cli.ai_handler.AIHandler.single_question" href="#connpy.cli.ai_handler.AIHandler.single_question">single_question</a></code></li>
@@ -367,7 +608,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.api_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -193,7 +193,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.config_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -482,7 +482,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.context_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -249,7 +249,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.forms API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -249,11 +249,183 @@ el.replaceWith(d);
        if &#34;tags&#34; in answer and not answer[&#34;tags&#34;].startswith(&#34;@&#34;) and answer[&#34;tags&#34;]:
            answer[&#34;tags&#34;] = ast.literal_eval(answer[&#34;tags&#34;])
-        return answer</code></pre>
+        return answer
    def mcp_wizard(self, mcp_servers):
        &#34;&#34;&#34;Interactive wizard to manage MCP servers.&#34;&#34;&#34;
        from .helpers import theme
        while True:
            options = [
                (&#34;List Configured Servers&#34;, &#34;list&#34;),
                (&#34;Add/Update Server&#34;, &#34;add&#34;),
                (&#34;Enable/Disable Server&#34;, &#34;toggle&#34;),
                (&#34;Remove Server&#34;, &#34;remove&#34;),
                (&#34;Back&#34;, &#34;exit&#34;)
            ]
            questions = [
                inquirer.List(&#34;action&#34;, message=&#34;MCP Configuration&#34;, choices=options)
            ]
            answers = inquirer.prompt(questions, theme=theme)
            if not answers or answers[&#34;action&#34;] == &#34;exit&#34;:
                return None
            action = answers[&#34;action&#34;]
            if action == &#34;list&#34;:
                if not mcp_servers:
                    print(&#34;\nNo MCP servers configured.\n&#34;)
                else:
                    return {&#34;action&#34;: &#34;list&#34;}
            elif action == &#34;add&#34;:
                questions = [
                    inquirer.Text(&#34;name&#34;, message=&#34;Server Name (identifier)&#34;),
                    inquirer.Text(&#34;url&#34;, message=&#34;SSE URL (e.g., http://localhost:8000/sse)&#34;),
                    inquirer.Confirm(&#34;enabled&#34;, message=&#34;Enabled?&#34;, default=True),
                    inquirer.Text(&#34;auto_load_os&#34;, message=&#34;Auto-load on specific OS (blank for any)&#34;)
                ]
                answers = inquirer.prompt(questions, theme=theme)
                if answers:
                    return {
                        &#34;action&#34;: &#34;add&#34;,
                        &#34;name&#34;: answers[&#34;name&#34;],
                        &#34;url&#34;: answers[&#34;url&#34;],
                        &#34;enabled&#34;: answers[&#34;enabled&#34;],
                        &#34;os&#34;: answers[&#34;auto_load_os&#34;]
                    }
            elif action == &#34;toggle&#34;:
                if not mcp_servers:
                    print(&#34;\nNo servers to toggle.\n&#34;)
                    continue
                choices = []
                for name, cfg in mcp_servers.items():
                    status = &#34;[Enabled]&#34; if cfg.get(&#34;enabled&#34;, True) else &#34;[Disabled]&#34;
                    choices.append((f&#34;{name} {status}&#34;, name))
                questions = [
                    inquirer.List(&#34;name&#34;, message=&#34;Select server to toggle&#34;, choices=choices + [(&#34;Cancel&#34;, None)])
                ]
                answers = inquirer.prompt(questions, theme=theme)
                if answers and answers[&#34;name&#34;]:
                    current = mcp_servers[answers[&#34;name&#34;]].get(&#34;enabled&#34;, True)
                    return {
                        &#34;action&#34;: &#34;update&#34;,
                        &#34;name&#34;: answers[&#34;name&#34;],
                        &#34;enabled&#34;: not current
                    }
            elif action == &#34;remove&#34;:
                if not mcp_servers:
                    print(&#34;\nNo servers to remove.\n&#34;)
                    continue
                questions = [
                    inquirer.List(&#34;name&#34;, message=&#34;Select server to remove&#34;, choices=list(mcp_servers.keys()) + [&#34;Cancel&#34;])
                ]
                answers = inquirer.prompt(questions, theme=theme)
                if answers and answers[&#34;name&#34;] != &#34;Cancel&#34;:
                    return {&#34;action&#34;: &#34;remove&#34;, &#34;name&#34;: answers[&#34;name&#34;]}
        return None</code></pre>
 </details>
 <div class="desc"></div>
 <h3>Methods</h3>
 <dl>
 <dt id="connpy.cli.forms.Forms.mcp_wizard"><code class="name flex">
 <span>def <span class="ident">mcp_wizard</span></span>(<span>self, mcp_servers)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def mcp_wizard(self, mcp_servers):
    &#34;&#34;&#34;Interactive wizard to manage MCP servers.&#34;&#34;&#34;
    from .helpers import theme
    while True:
        options = [
            (&#34;List Configured Servers&#34;, &#34;list&#34;),
            (&#34;Add/Update Server&#34;, &#34;add&#34;),
            (&#34;Enable/Disable Server&#34;, &#34;toggle&#34;),
            (&#34;Remove Server&#34;, &#34;remove&#34;),
            (&#34;Back&#34;, &#34;exit&#34;)
        ]
        questions = [
            inquirer.List(&#34;action&#34;, message=&#34;MCP Configuration&#34;, choices=options)
        ]
        answers = inquirer.prompt(questions, theme=theme)
        if not answers or answers[&#34;action&#34;] == &#34;exit&#34;:
            return None
        action = answers[&#34;action&#34;]
        if action == &#34;list&#34;:
            if not mcp_servers:
                print(&#34;\nNo MCP servers configured.\n&#34;)
            else:
                return {&#34;action&#34;: &#34;list&#34;}
        elif action == &#34;add&#34;:
            questions = [
                inquirer.Text(&#34;name&#34;, message=&#34;Server Name (identifier)&#34;),
                inquirer.Text(&#34;url&#34;, message=&#34;SSE URL (e.g., http://localhost:8000/sse)&#34;),
                inquirer.Confirm(&#34;enabled&#34;, message=&#34;Enabled?&#34;, default=True),
                inquirer.Text(&#34;auto_load_os&#34;, message=&#34;Auto-load on specific OS (blank for any)&#34;)
            ]
            answers = inquirer.prompt(questions, theme=theme)
            if answers:
                return {
                    &#34;action&#34;: &#34;add&#34;,
                    &#34;name&#34;: answers[&#34;name&#34;],
                    &#34;url&#34;: answers[&#34;url&#34;],
                    &#34;enabled&#34;: answers[&#34;enabled&#34;],
                    &#34;os&#34;: answers[&#34;auto_load_os&#34;]
                }
        elif action == &#34;toggle&#34;:
            if not mcp_servers:
                print(&#34;\nNo servers to toggle.\n&#34;)
                continue
            choices = []
            for name, cfg in mcp_servers.items():
                status = &#34;[Enabled]&#34; if cfg.get(&#34;enabled&#34;, True) else &#34;[Disabled]&#34;
                choices.append((f&#34;{name} {status}&#34;, name))
            questions = [
                inquirer.List(&#34;name&#34;, message=&#34;Select server to toggle&#34;, choices=choices + [(&#34;Cancel&#34;, None)])
            ]
            answers = inquirer.prompt(questions, theme=theme)
            if answers and answers[&#34;name&#34;]:
                current = mcp_servers[answers[&#34;name&#34;]].get(&#34;enabled&#34;, True)
                return {
                    &#34;action&#34;: &#34;update&#34;,
                    &#34;name&#34;: answers[&#34;name&#34;],
                    &#34;enabled&#34;: not current
                }
        elif action == &#34;remove&#34;:
            if not mcp_servers:
                print(&#34;\nNo servers to remove.\n&#34;)
                continue
            questions = [
                inquirer.List(&#34;name&#34;, message=&#34;Select server to remove&#34;, choices=list(mcp_servers.keys()) + [&#34;Cancel&#34;])
            ]
            answers = inquirer.prompt(questions, theme=theme)
            if answers and answers[&#34;name&#34;] != &#34;Cancel&#34;:
                return {&#34;action&#34;: &#34;remove&#34;, &#34;name&#34;: answers[&#34;name&#34;]}
    return None</code></pre>
 </details>
 <div class="desc"><p>Interactive wizard to manage MCP servers.</p></div>
 </dd>
 <dt id="connpy.cli.forms.Forms.questions_bulk"><code class="name flex">
 <span>def <span class="ident">questions_bulk</span></span>(<span>self, nodes='', hosts='')</span>
 </code></dt>
@@ -505,6 +677,7 @@ el.replaceWith(d);
 <li>
 <h4><code><a title="connpy.cli.forms.Forms" href="#connpy.cli.forms.Forms">Forms</a></code></h4>
 <ul class="">
 <li><code><a title="connpy.cli.forms.Forms.mcp_wizard" href="#connpy.cli.forms.Forms.mcp_wizard">mcp_wizard</a></code></li>
 <li><code><a title="connpy.cli.forms.Forms.questions_bulk" href="#connpy.cli.forms.Forms.questions_bulk">questions_bulk</a></code></li>
 <li><code><a title="connpy.cli.forms.Forms.questions_edit" href="#connpy.cli.forms.Forms.questions_edit">questions_edit</a></code></li>
 <li><code><a title="connpy.cli.forms.Forms.questions_nodes" href="#connpy.cli.forms.Forms.questions_nodes">questions_nodes</a></code></li>
@@ -517,7 +690,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.help_text API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -303,7 +303,7 @@ tasks:
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.helpers API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -207,7 +207,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.import_export_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -272,7 +272,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -92,6 +92,10 @@ el.replaceWith(d);
 <dd>
 <div class="desc"></div>
 </dd>
 <dt><code class="name"><a title="connpy.cli.terminal_ui" href="terminal_ui.html">connpy.cli.terminal_ui</a></code></dt>
 <dd>
 <div class="desc"></div>
 </dd>
 <dt><code class="name"><a title="connpy.cli.validators" href="validators.html">connpy.cli.validators</a></code></dt>
 <dd>
 <div class="desc"></div>
@@ -130,6 +134,7 @@ el.replaceWith(d);
 <li><code><a title="connpy.cli.profile_handler" href="profile_handler.html">connpy.cli.profile_handler</a></code></li>
 <li><code><a title="connpy.cli.run_handler" href="run_handler.html">connpy.cli.run_handler</a></code></li>
 <li><code><a title="connpy.cli.sync_handler" href="sync_handler.html">connpy.cli.sync_handler</a></code></li>
 <li><code><a title="connpy.cli.terminal_ui" href="terminal_ui.html">connpy.cli.terminal_ui</a></code></li>
 <li><code><a title="connpy.cli.validators" href="validators.html">connpy.cli.validators</a></code></li>
 </ul>
 </li>
@@ -137,7 +142,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.node_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -606,7 +606,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.plugin_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -385,7 +385,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.profile_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -314,7 +314,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.run_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -169,7 +169,7 @@ el.replaceWith(d);
                    commands=commands,
                    variables=variables,
                    parallel=options.get(&#34;parallel&#34;, 10),
-                    timeout=options.get(&#34;timeout&#34;, 10),
+                    timeout=options.get(&#34;timeout&#34;, 20),
                    folder=folder,
                    prompt=prompt,
                    on_node_complete=_on_run_complete
@@ -203,7 +203,7 @@ el.replaceWith(d);
                    expected=expected,
                    variables=variables,
                    parallel=options.get(&#34;parallel&#34;, 10),
-                    timeout=options.get(&#34;timeout&#34;, 10),
+                    timeout=options.get(&#34;timeout&#34;, 20),
                    folder=folder,
                    prompt=prompt,
                    on_node_complete=_on_test_complete
@@ -260,7 +260,7 @@ el.replaceWith(d);
                commands=commands,
                variables=variables,
                parallel=options.get(&#34;parallel&#34;, 10),
-                timeout=options.get(&#34;timeout&#34;, 10),
+                timeout=options.get(&#34;timeout&#34;, 20),
                folder=folder,
                prompt=prompt,
                on_node_complete=_on_run_complete
@@ -294,7 +294,7 @@ el.replaceWith(d);
                expected=expected,
                variables=variables,
                parallel=options.get(&#34;parallel&#34;, 10),
-                timeout=options.get(&#34;timeout&#34;, 10),
+                timeout=options.get(&#34;timeout&#34;, 20),
                folder=folder,
                prompt=prompt,
                on_node_complete=_on_test_complete
@@ -454,7 +454,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.sync_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -427,7 +427,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.cli.validators API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -508,7 +508,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.grpc_layer.connpy_pb2 API documentation</title>
 <meta name="description" content="Generated protocol buffer code.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -62,7 +62,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.AIResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -81,7 +81,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.AskRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -100,7 +100,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.BoolResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -119,7 +119,45 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.BulkRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
 <dt id="connpy.grpc_layer.connpy_pb2.CopilotRequest"><code class="flex name class">
 <span>class <span class="ident">CopilotRequest</span></span>
 <span>(</span><span>*args, **kwargs)</span>
 </code></dt>
 <dd>
 <div class="desc"><p>A ProtocolMessage</p></div>
 <h3>Ancestors</h3>
 <ul class="hlist">
 <li>google._upb._message.Message</li>
 <li>google.protobuf.message.Message</li>
 </ul>
 <h3>Class variables</h3>
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.CopilotRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
 <div class="desc"></div>
 </dd>
 </dl>
 </dd>
 <dt id="connpy.grpc_layer.connpy_pb2.CopilotResponse"><code class="flex name class">
 <span>class <span class="ident">CopilotResponse</span></span>
 <span>(</span><span>*args, **kwargs)</span>
 </code></dt>
 <dd>
 <div class="desc"><p>A ProtocolMessage</p></div>
 <h3>Ancestors</h3>
 <ul class="hlist">
 <li>google._upb._message.Message</li>
 <li>google.protobuf.message.Message</li>
 </ul>
 <h3>Class variables</h3>
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.CopilotResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
 <div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -138,7 +176,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.DeleteRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -157,7 +195,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.ExportRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -176,7 +214,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.FilterRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -195,7 +233,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.FullReplaceRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -214,7 +252,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.IdRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -233,7 +271,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.IntRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -252,7 +290,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.InteractRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -271,7 +309,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.InteractResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -290,7 +328,26 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.ListRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
 <dt id="connpy.grpc_layer.connpy_pb2.MCPRequest"><code class="flex name class">
 <span>class <span class="ident">MCPRequest</span></span>
 <span>(</span><span>*args, **kwargs)</span>
 </code></dt>
 <dd>
 <div class="desc"><p>A ProtocolMessage</p></div>
 <h3>Ancestors</h3>
 <ul class="hlist">
 <li>google._upb._message.Message</li>
 <li>google.protobuf.message.Message</li>
 </ul>
 <h3>Class variables</h3>
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.MCPRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
 <div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -309,7 +366,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.MessageValue.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -328,7 +385,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.MoveRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -347,7 +404,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.NodeRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -366,7 +423,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.NodeRunResult.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -385,7 +442,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.PluginRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -404,7 +461,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.ProfileRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -423,7 +480,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.ProviderRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -442,7 +499,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.RunRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -461,7 +518,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.ScriptRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -480,7 +537,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.StringRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -499,7 +556,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.StringResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -518,7 +575,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.StructRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -537,7 +594,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.StructResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -556,7 +613,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.TestRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -575,7 +632,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.UpdateRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -594,7 +651,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.connpy_pb2.ValueResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -638,6 +695,18 @@ el.replaceWith(d);
 </ul>
 </li>
 <li>
 <h4><code><a title="connpy.grpc_layer.connpy_pb2.CopilotRequest" href="#connpy.grpc_layer.connpy_pb2.CopilotRequest">CopilotRequest</a></code></h4>
 <ul class="">
 <li><code><a title="connpy.grpc_layer.connpy_pb2.CopilotRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.CopilotRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
 </ul>
 </li>
 <li>
 <h4><code><a title="connpy.grpc_layer.connpy_pb2.CopilotResponse" href="#connpy.grpc_layer.connpy_pb2.CopilotResponse">CopilotResponse</a></code></h4>
 <ul class="">
 <li><code><a title="connpy.grpc_layer.connpy_pb2.CopilotResponse.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.CopilotResponse.DESCRIPTOR">DESCRIPTOR</a></code></li>
 </ul>
 </li>
 <li>
 <h4><code><a title="connpy.grpc_layer.connpy_pb2.DeleteRequest" href="#connpy.grpc_layer.connpy_pb2.DeleteRequest">DeleteRequest</a></code></h4>
 <ul class="">
 <li><code><a title="connpy.grpc_layer.connpy_pb2.DeleteRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.DeleteRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
@@ -692,6 +761,12 @@ el.replaceWith(d);
 </ul>
 </li>
 <li>
 <h4><code><a title="connpy.grpc_layer.connpy_pb2.MCPRequest" href="#connpy.grpc_layer.connpy_pb2.MCPRequest">MCPRequest</a></code></h4>
 <ul class="">
 <li><code><a title="connpy.grpc_layer.connpy_pb2.MCPRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.MCPRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
 </ul>
 </li>
 <li>
 <h4><code><a title="connpy.grpc_layer.connpy_pb2.MessageValue" href="#connpy.grpc_layer.connpy_pb2.MessageValue">MessageValue</a></code></h4>
 <ul class="">
 <li><code><a title="connpy.grpc_layer.connpy_pb2.MessageValue.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.MessageValue.DESCRIPTOR">DESCRIPTOR</a></code></li>
@@ -793,7 +868,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.grpc_layer API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -102,7 +102,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.grpc_layer.remote_plugin_pb2 API documentation</title>
 <meta name="description" content="Generated protocol buffer code.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -62,7 +62,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.remote_plugin_pb2.IdRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -81,7 +81,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.remote_plugin_pb2.OutputChunk.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -100,7 +100,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.remote_plugin_pb2.PluginInvokeRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -119,7 +119,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.remote_plugin_pb2.StringResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"><p>The type of the None singleton.</p></div>
+<div class="desc"></div>
 </dd>
 </dl>
 </dd>
@@ -168,7 +168,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.grpc_layer.remote_plugin_pb2_grpc API documentation</title>
 <meta name="description" content="Client and server classes corresponding to protobuf-defined services.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -366,7 +366,7 @@ def invoke_plugin(request,
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.grpc_layer.server API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -96,7 +96,7 @@ el.replaceWith(d);
    interceptors = [LoggingInterceptor()] if debug else []
    server = grpc.server(futures.ThreadPoolExecutor(max_workers=10), interceptors=interceptors)
-    connpy_pb2_grpc.add_NodeServiceServicer_to_server(NodeServicer(config), server)
+    connpy_pb2_grpc.add_NodeServiceServicer_to_server(NodeServicer(config, debug=debug), server)
    connpy_pb2_grpc.add_ProfileServiceServicer_to_server(ProfileServicer(config), server)
    connpy_pb2_grpc.add_ConfigServiceServicer_to_server(ConfigServicer(config), server)
    plugin_servicer = PluginServicer(config)
@@ -245,6 +245,22 @@ el.replaceWith(d);
        res = self.service.confirm(request.value)
        return connpy_pb2.BoolResponse(value=res)
    @handle_errors
    def ask_copilot(self, request, context):
        import json
        node_info = json.loads(request.node_info_json) if request.node_info_json else None
        result = self.service.ask_copilot(
            request.terminal_buffer,
            request.user_question,
            node_info
        )
        return connpy_pb2.CopilotResponse(
            commands=result.get(&#34;commands&#34;, []),
            guide=result.get(&#34;guide&#34;, &#34;&#34;),
            risk_level=result.get(&#34;risk_level&#34;, &#34;low&#34;),
            error=result.get(&#34;error&#34;) or &#34;&#34;
        )
    @handle_errors
    def list_sessions(self, request, context):
        return connpy_pb2.ValueResponse(data=to_value(self.service.list_sessions()))
@@ -259,6 +275,17 @@ el.replaceWith(d);
        self.service.configure_provider(request.provider, request.model, request.api_key)
        return Empty()
    @handle_errors
    def configure_mcp(self, request, context):
        self.service.configure_mcp(
            request.name, 
            url=request.url or None, 
            enabled=request.enabled, 
            auto_load_on_os=request.auto_load_on_os or None, 
            remove=request.remove
        )
        return Empty()
    @handle_errors
    def load_session_data(self, request, context):
        return connpy_pb2.StructResponse(data=to_struct(self.service.load_session_data(request.value)))</code></pre>
@@ -273,6 +300,8 @@ el.replaceWith(d);
 <li><code><b><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer">AIServiceServicer</a></b></code>:
 <ul class="hlist">
 <li><code><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.ask" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.ask">ask</a></code></li>
 <li><code><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.ask_copilot" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.ask_copilot">ask_copilot</a></code></li>
 <li><code><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.configure_mcp" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.configure_mcp">configure_mcp</a></code></li>
 <li><code><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.configure_provider" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.configure_provider">configure_provider</a></code></li>
 <li><code><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.confirm" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.confirm">confirm</a></code></li>
 <li><code><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.delete_session" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.delete_session">delete_session</a></code></li>
@@ -618,7 +647,7 @@ interceptor chooses to service this RPC, or None otherwise.</p></div>
 </dd>
 <dt id="connpy.grpc_layer.server.NodeServicer"><code class="flex name class">
 <span>class <span class="ident">NodeServicer</span></span>
-<span>(</span><span>config)</span>
+<span>(</span><span>config, debug=False)</span>
 </code></dt>
 <dd>
 <details class="source">
@@ -626,8 +655,13 @@ interceptor chooses to service this RPC, or None otherwise.</p></div>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">class NodeServicer(connpy_pb2_grpc.NodeServiceServicer):
-    def __init__(self, config):
+    def __init__(self, config, debug=False):
        self.service = NodeService(config)
        self.server_debug = debug
        if debug:
            from rich.console import Console
            from ..printer import connpy_theme, get_original_stdout
            self.server_console = Console(theme=connpy_theme, file=get_original_stdout())
    @handle_errors
    def interact_node(self, request_iterator, context):
@@ -650,8 +684,8 @@ interceptor chooses to service this RPC, or None otherwise.</p></div>
        sftp = first_req.sftp
        debug = first_req.debug
-        if debug:
+        if self.server_debug:
-            printer.console.print(f&#34;[debug][DEBUG][/debug] gRPC interact_node request for: [bold cyan]{unique_id}[/bold cyan]&#34;)
+            self.server_console.print(f&#34;[debug][DEBUG][/debug] gRPC interact_node request for: [bold cyan]{unique_id}[/bold cyan]&#34;)
        if first_req.connection_params_json:
            import json
@@ -710,7 +744,39 @@ interceptor chooses to service this RPC, or None otherwise.</p></div>
            if sftp:
                n.protocol = &#34;sftp&#34;
-        connect = n._connect(debug=debug)
+        # Build a logger that captures debug messages as ANSI-colored bytes for the client
        debug_chunks = []
        if debug:
            from io import StringIO
            from rich.console import Console as RichConsole
            from ..printer import connpy_theme
            from .. import printer as _printer
            def remote_logger(msg_type, message):
                buf = StringIO()
                c = RichConsole(file=buf, force_terminal=True, width=120, theme=connpy_theme)
                if msg_type == &#34;debug&#34;:
                    c.print(_printer._format_multiline(&#34;i&#34;, f&#34;[DEBUG] {message}&#34;, style=&#34;info&#34;))
                elif msg_type == &#34;success&#34;:
                    c.print(_printer._format_multiline(&#34;✓&#34;, message, style=&#34;success&#34;))
                elif msg_type == &#34;error&#34;:
                    c.print(_printer._format_multiline(&#34;✗&#34;, message, style=&#34;error&#34;))
                else:
                    c.print(str(message))
                rendered = buf.getvalue()
                if rendered:
                    # Raw TTY needs \r\n instead of \n
                    rendered = rendered.replace(&#39;\n&#39;, &#39;\r\n&#39;)
                    debug_chunks.append(rendered.encode())
        else:
            remote_logger = None
        connect = n._connect(debug=debug, logger=remote_logger)
        # Send debug output to client before checking result (always show the command)
        for chunk in debug_chunks:
            yield connpy_pb2.InteractResponse(stdout_data=chunk)
        if connect != True:
            yield connpy_pb2.InteractResponse(success=False, error_message=str(connect))
            return
@@ -737,7 +803,177 @@ interceptor chooses to service this RPC, or None otherwise.</p></div>
                    except Exception:
                        pass
-                asyncio.run(n._async_interact_loop(remote_stream, resize_callback))
+                async def remote_copilot_handler(buffer, node_info, stream, child_fd, cmd_byte_positions=None):
                    import json
                    import asyncio
                    import os
                    from ..services.ai_service import AIService
                    service = AIService(self.service.config)
                    if node_info is None:
                        node_info = {}
                    # Calculate real command blocks from history using the central service
                    raw_bytes = n.mylog.getvalue() if hasattr(n, &#39;mylog&#39;) else buffer
                    if not isinstance(raw_bytes, bytes):
                        raw_bytes = str(raw_bytes).encode()
                    from connpy.utils import log_cleaner
                    last_line = log_cleaner(raw_bytes.decode(errors=&#39;replace&#39;)).split(&#39;\n&#39;)[-1].strip()
                    blocks = service.build_context_blocks(raw_bytes, n.cmd_byte_positions, node_info, last_line=last_line)
                    node_info[&#34;context_blocks&#34;] = blocks
                    node_info_json = json.dumps(node_info)
                    # Convert buffer to string if it&#39;s bytes for the preview
                    preview_str = buffer[-200:].decode(errors=&#39;replace&#39;) if isinstance(buffer, bytes) else str(buffer)[-200:]
                    # Generate a unique session ID for this copilot interaction to prevent race conditions
                    import uuid
                    copilot_session_id = str(uuid.uuid4())
                    node_info[&#34;session_id&#34;] = copilot_session_id
                    node_info_json = json.dumps(node_info)
                    # 1. Send prompt to client
                    response_queue.put(connpy_pb2.InteractResponse(
                        copilot_prompt=True,
                        copilot_buffer_preview=preview_str,
                        copilot_node_info_json=node_info_json
                    ))
                    while True:
                        # 0. Drain the queue of any stale messages before starting a new interaction
                        while not remote_stream.copilot_queue.empty():
                            try:
                                remote_stream.copilot_queue.get_nowait()
                            except:
                                break
                        # 2. Await the question from client via the copilot_queue
                        import threading
                        def preload_ai_deps():
                            try:
                                import litellm
                            except Exception:
                                pass
                        threading.Thread(target=preload_ai_deps, daemon=True).start()
                        try:
                            req_data = await asyncio.wait_for(remote_stream.copilot_queue.get(), timeout=120)
                            if not req_data: return
                            # Validate session ID if provided by client (skip validation if not provided for CLI compatibility)
                            req_session_id = req_data.get(&#34;session_id&#34;)
                            if req_session_id and req_session_id != copilot_session_id:
                                continue # Ignore stale request from a previous session
                            if &#34;question&#34; not in req_data or not req_data[&#34;question&#34;] or req_data[&#34;question&#34;] == &#34;CANCEL&#34; or req_data.get(&#34;action&#34;) in (&#34;cancel&#34;, &#34;web_cancel&#34;):
                                if req_data.get(&#34;action&#34;) == &#34;web_cancel&#34;:
                                    os.write(child_fd, b&#39;\x05&#39;)
                                else:
                                    os.write(child_fd, b&#39;\x15\r&#39;)
                                return
                            question = req_data[&#34;question&#34;]
                            merged_node_info_str = req_data.get(&#34;node_info_json&#34;, &#34;&#34;)
                            if merged_node_info_str:
                                try:
                                    merged_node_info = json.loads(merged_node_info_str)
                                    node_info.update(merged_node_info)
                                except: pass
                            context_buffer = req_data.get(&#34;context_buffer&#34;, &#34;&#34;)
                            if context_buffer.startswith(&#39;{&#34;context_start_pos&#34;&#39;):
                                try:
                                    parsed = json.loads(context_buffer)
                                    start_pos = parsed[&#34;context_start_pos&#34;]
                                    selected_raw = raw_bytes[start_pos:]
                                    context_buffer = n._logclean(selected_raw.decode(errors=&#39;replace&#39;), var=True)
                                except Exception:
                                    context_buffer = buffer
                            elif not context_buffer:
                                context_buffer = buffer
                        except asyncio.TimeoutError:
                            os.write(child_fd, b&#39;\x15\r&#39;)
                            return
                        # 3. Call AI Service with streaming
                        def chunk_callback(chunk_text):
                            if chunk_text:
                                response_queue.put(connpy_pb2.InteractResponse(
                                    copilot_stream_chunk=chunk_text
                                ))
                        # Create a clean version of node_info for the AI to save tokens and match local CLI behavior
                        ai_node_info = {k: v for k, v in node_info.items() if k not in (&#34;context_blocks&#34;, &#34;full_buffer&#34;)}
                        ai_task = asyncio.create_task(service.aask_copilot(context_buffer, question, ai_node_info, chunk_callback=chunk_callback))
                        wait_action_task = asyncio.create_task(remote_stream.copilot_queue.get())
                        done, pending = await asyncio.wait(
                            [ai_task, wait_action_task],
                            return_when=asyncio.FIRST_COMPLETED
                        )
                        if wait_action_task in done:
                            req_data = wait_action_task.result()
                            ai_task.cancel()
                            if req_data.get(&#34;action&#34;) in (&#34;cancel&#34;, &#34;web_cancel&#34;) or req_data.get(&#34;question&#34;) == &#34;CANCEL&#34;:
                                if req_data.get(&#34;action&#34;) == &#34;web_cancel&#34;:
                                    os.write(child_fd, b&#39;\x05&#39;)
                                else:
                                    os.write(child_fd, b&#39;\x15\r&#39;)
                                return
                            continue # Loop back instead of returning to keep session alive
                        else:
                            wait_action_task.cancel()
                            result = ai_task.result()
                            if not result:
                                os.write(child_fd, b&#39;\x15\r&#39;)
                                return
                        # 4. Send response back to client
                        response_queue.put(connpy_pb2.InteractResponse(
                            copilot_response_json=json.dumps(result)
                        ))
                        # 5. Wait for user action
                        try:
                            action_data = await asyncio.wait_for(remote_stream.copilot_queue.get(), timeout=60)
                            if not action_data: return
                            action = action_data.get(&#34;action&#34;, &#34;cancel&#34;)
                            if action == &#34;continue&#34;:
                                continue # Loop back for next question
                            if action in (&#34;cancel&#34;, &#34;web_cancel&#34;):
                                if action == &#34;web_cancel&#34;:
                                    os.write(child_fd, b&#39;\x05&#39;)
                                else:
                                    os.write(child_fd, b&#39;\x15\r&#39;)
                                return
                        except asyncio.TimeoutError:
                            os.write(child_fd, b&#39;\x15\r&#39;)
                            return
                        def on_inject(cmd):
                            response_queue.put(connpy_pb2.InteractResponse(copilot_injected_command=cmd))
                        if action == &#34;send_all&#34;:
                            commands = result.get(&#34;commands&#34;, [])
                            await n.inject_commands(commands, child_fd, on_inject=on_inject)
                            return
                        elif action.startswith(&#34;custom:&#34;):
                            custom_cmds_raw = action[7:]
                            custom_cmds = [cmd.strip() for cmd in custom_cmds_raw.split(&#39;\n&#39;) if cmd.strip()]
                            await n.inject_commands(custom_cmds, child_fd, on_inject=on_inject)
                            return
                        else:
                            os.write(child_fd, b&#39;\x15\r&#39;)
                            return
                asyncio.run(n._async_interact_loop(remote_stream, resize_callback, copilot_handler=remote_copilot_handler))
            except Exception as e:
                pass
            finally:
@@ -746,14 +982,19 @@ interceptor chooses to service this RPC, or None otherwise.</p></div>
        t_loop = threading.Thread(target=run_async_loop, daemon=True)
        t_loop.start()
-
+        def response_generator():
            while True:
                data = response_queue.get()
                if data is None:
-                if debug:
+                    if self.server_debug:
-                    printer.console.print(f&#34;[debug][DEBUG][/debug] gRPC interact_node session closed for: [bold cyan]{unique_id}[/bold cyan]&#34;)
+                        self.server_console.print(f&#34;[debug][DEBUG][/debug] gRPC interact_node session closed for: [bold cyan]{unique_id}[/bold cyan]&#34;)
                    break
                if isinstance(data, connpy_pb2.InteractResponse):
                    yield data
                else:
                    yield connpy_pb2.InteractResponse(stdout_data=data)
        yield from response_generator()
    @handle_errors
    def list_nodes(self, request, context):
        f = request.filter_str if request.filter_str else None
@@ -1319,7 +1560,7 @@ interceptor chooses to service this RPC, or None otherwise.</p></div>
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.grpc_layer.stubs API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -104,7 +104,6 @@ el.replaceWith(d);
        import queue
        from rich.prompt import Prompt
        from rich.text import Text
        from rich.live import Live
        from rich.panel import Panel
        from rich.markdown import Markdown
@@ -135,7 +134,7 @@ el.replaceWith(d);
        responses = self.stub.ask(request_generator())
        full_content = &#34;&#34;
-        live_display = None
+        header_printed = False
        final_result = {&#34;response&#34;: &#34;&#34;, &#34;chat_history&#34;: []}
        # Background thread to pull responses from gRPC into a local queue
@@ -221,17 +220,32 @@ el.replaceWith(d);
                if not response.is_final:
                    if response.text_chunk:
-                        full_content += response.text_chunk
+                        if not header_printed:
                            if status:
                                try: status.stop()
                                except: pass
-                        if status and not debug:
+                            from rich.console import Console as RichConsole
-                            # Update the spinner line with a preview of the response
+                            from rich.rule import Rule
-                            preview = full_content.replace(&#34;\n&#34;, &#34; &#34;).strip()
+                            from ..printer import connpy_theme, get_original_stdout, IncrementalMarkdownParser
-                            if len(preview) &gt; 60: preview = preview[:57] + &#34;...&#34;
+                            stable_console = RichConsole(theme=connpy_theme, file=get_original_stdout())
-                            status.update(f&#34;[ai_status]{preview}&#34;)
+                            
                            # Print header on first chunk
                            stable_console.print(Rule(&#34;[bold engineer]Network Engineer[/bold engineer]&#34;, style=&#34;engineer&#34;))
                            header_printed = True
                            # Initialize parser
                            md_parser = IncrementalMarkdownParser(console=stable_console)
                        full_content += response.text_chunk
                        md_parser.feed(response.text_chunk)
                    continue
                if response.is_final:
-                    # Final stop for status to ensure it disappears before the panel
+                    if header_printed:
                        from rich.rule import Rule
                        md_parser.flush()
                    if status:
                        try: status.stop()
                        except: pass
@@ -242,10 +256,14 @@ el.replaceWith(d);
                    role_label = &#34;Network Architect&#34; if responder == &#34;architect&#34; else &#34;Network Engineer&#34;
                    title = f&#34;[bold {alias}]{role_label}[/bold {alias}]&#34;
-                    # Always print the final Panel
+                    if header_printed:
-                    content_to_print = full_content or final_result.get(&#34;response&#34;, &#34;&#34;)
+                        from rich.console import Console as RichConsole
-                    if content_to_print:
+                        from ..printer import connpy_theme, get_original_stdout
-                        printer.console.print(Panel(Markdown(content_to_print), title=title, border_style=alias, expand=False))
+                        stable_console = RichConsole(theme=connpy_theme, file=get_original_stdout())
                        stable_console.print(Rule(style=alias))
                    elif not full_content and final_result.get(&#34;response&#34;):
                        # If nothing streamed but we have response (e.g. error or direct guide)
                        printer.console.print(Panel(Markdown(final_result[&#34;response&#34;]), title=title, border_style=alias, expand=False))
                    break
        except Exception as e:
            # Check if it was a gRPC error that we should let handle_errors catch
@@ -277,6 +295,17 @@ el.replaceWith(d);
        req = connpy_pb2.ProviderRequest(provider=provider, model=model or &#34;&#34;, api_key=api_key or &#34;&#34;)
        self.stub.configure_provider(req)
    @handle_errors
    def configure_mcp(self, name, url=None, enabled=True, auto_load_on_os=None, remove=False):
        req = connpy_pb2.MCPRequest(
            name=name,
            url=url or &#34;&#34;,
            enabled=enabled,
            auto_load_on_os=auto_load_on_os or &#34;&#34;,
            remove=remove
        )
        self.stub.configure_mcp(req)
    @handle_errors
    def load_session_data(self, session_id):
        return from_struct(self.stub.load_session_data(connpy_pb2.StringRequest(value=session_id)).data)</code></pre>
@@ -297,7 +326,6 @@ def ask(self, input_text, dryrun=False, chat_history=None, session_id=None, debu
    import queue
    from rich.prompt import Prompt
    from rich.text import Text
    from rich.live import Live
    from rich.panel import Panel
    from rich.markdown import Markdown
@@ -328,7 +356,7 @@ def ask(self, input_text, dryrun=False, chat_history=None, session_id=None, debu
    responses = self.stub.ask(request_generator())
    full_content = &#34;&#34;
-    live_display = None
+    header_printed = False
    final_result = {&#34;response&#34;: &#34;&#34;, &#34;chat_history&#34;: []}
    # Background thread to pull responses from gRPC into a local queue
@@ -414,17 +442,32 @@ def ask(self, input_text, dryrun=False, chat_history=None, session_id=None, debu
            if not response.is_final:
                if response.text_chunk:
-                    full_content += response.text_chunk
+                    if not header_printed:
                        if status:
                            try: status.stop()
                            except: pass
-                    if status and not debug:
+                        from rich.console import Console as RichConsole
-                        # Update the spinner line with a preview of the response
+                        from rich.rule import Rule
-                        preview = full_content.replace(&#34;\n&#34;, &#34; &#34;).strip()
+                        from ..printer import connpy_theme, get_original_stdout, IncrementalMarkdownParser
-                        if len(preview) &gt; 60: preview = preview[:57] + &#34;...&#34;
+                        stable_console = RichConsole(theme=connpy_theme, file=get_original_stdout())
-                        status.update(f&#34;[ai_status]{preview}&#34;)
+                        
                        # Print header on first chunk
                        stable_console.print(Rule(&#34;[bold engineer]Network Engineer[/bold engineer]&#34;, style=&#34;engineer&#34;))
                        header_printed = True
                        # Initialize parser
                        md_parser = IncrementalMarkdownParser(console=stable_console)
                    full_content += response.text_chunk
                    md_parser.feed(response.text_chunk)
                continue
            if response.is_final:
-                # Final stop for status to ensure it disappears before the panel
+                if header_printed:
                    from rich.rule import Rule
                    md_parser.flush()
                if status:
                    try: status.stop()
                    except: pass
@@ -435,10 +478,14 @@ def ask(self, input_text, dryrun=False, chat_history=None, session_id=None, debu
                role_label = &#34;Network Architect&#34; if responder == &#34;architect&#34; else &#34;Network Engineer&#34;
                title = f&#34;[bold {alias}]{role_label}[/bold {alias}]&#34;
-                # Always print the final Panel
+                if header_printed:
-                content_to_print = full_content or final_result.get(&#34;response&#34;, &#34;&#34;)
+                    from rich.console import Console as RichConsole
-                if content_to_print:
+                    from ..printer import connpy_theme, get_original_stdout
-                    printer.console.print(Panel(Markdown(content_to_print), title=title, border_style=alias, expand=False))
+                    stable_console = RichConsole(theme=connpy_theme, file=get_original_stdout())
                    stable_console.print(Rule(style=alias))
                elif not full_content and final_result.get(&#34;response&#34;):
                    # If nothing streamed but we have response (e.g. error or direct guide)
                    printer.console.print(Panel(Markdown(final_result[&#34;response&#34;]), title=title, border_style=alias, expand=False))
                break
    except Exception as e:
        # Check if it was a gRPC error that we should let handle_errors catch
@@ -455,6 +502,27 @@ def ask(self, input_text, dryrun=False, chat_history=None, session_id=None, debu
 </details>
 <div class="desc"></div>
 </dd>
 <dt id="connpy.grpc_layer.stubs.AIStub.configure_mcp"><code class="name flex">
 <span>def <span class="ident">configure_mcp</span></span>(<span>self, name, url=None, enabled=True, auto_load_on_os=None, remove=False)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">@handle_errors
 def configure_mcp(self, name, url=None, enabled=True, auto_load_on_os=None, remove=False):
    req = connpy_pb2.MCPRequest(
        name=name,
        url=url or &#34;&#34;,
        enabled=enabled,
        auto_load_on_os=auto_load_on_os or &#34;&#34;,
        remove=remove
    )
    self.stub.configure_mcp(req)</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
 <dt id="connpy.grpc_layer.stubs.AIStub.configure_provider"><code class="name flex">
 <span>def <span class="ident">configure_provider</span></span>(<span>self, provider, model=None, api_key=None)</span>
 </code></dt>
@@ -924,15 +992,100 @@ def set_reserved_names(self, names):
        self.remote_host = remote_host
        self.config = config
    def _handle_remote_copilot(self, res, request_queue, response_queue, client_buffer_bytes, pause_generator, resume_generator, old_tty):
        import json, asyncio, termios, sys, tty, queue
        from ..core import copilot_terminal_mode
        from . import connpy_pb2
        pause_generator()
        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
        node_info = json.loads(res.copilot_node_info_json) if res.copilot_node_info_json else {}
        blocks = node_info.get(&#34;context_blocks&#34;, [])
        interface = CopilotInterface(
            self.config, 
            history=getattr(self, &#39;copilot_history&#39;, None),
            session_state=getattr(self, &#39;copilot_state&#39;, None)
        )
        self.copilot_history = interface.history
        self.copilot_state = interface.session_state
        async def on_ai_call_remote(active_buffer, question, chunk_callback, merged_node_info):
            # Send request to server
            request_queue.put(connpy_pb2.InteractRequest(
                copilot_question=question, 
                copilot_context_buffer=active_buffer,
                copilot_node_info_json=json.dumps(merged_node_info)
            ))
            # Wait for chunks from server
            while True:
                try:
                    chunk_res = response_queue.get(timeout=0.1)
                    if chunk_res is None: return {&#34;error&#34;: &#34;Server disconnected&#34;}
                    if chunk_res.copilot_stream_chunk:
                        chunk_callback(chunk_res.copilot_stream_chunk)
                    elif chunk_res.copilot_response_json:
                        return json.loads(chunk_res.copilot_response_json)
                except queue.Empty:
                    await asyncio.sleep(0.05)
        # Wrap in async loop
        async def run_remote_copilot():
            while True:
                action, commands, custom_cmd = await interface.run_session(
                    raw_bytes=bytes(client_buffer_bytes),
                    node_info=node_info,
                    on_ai_call=on_ai_call_remote,
                    blocks=blocks
                )
                if action == &#34;continue&#34;:
                    # Send continue signal to server to loop back for another question
                    request_queue.put(connpy_pb2.InteractRequest(copilot_action=&#34;continue&#34;))
                    continue
                return action, commands, custom_cmd
        with copilot_terminal_mode():
            action, commands, custom_cmd = asyncio.run(run_remote_copilot())
        print(&#34;\033[2m Returning to session...\033[0m&#34;, flush=True)
        # Prepare final action for server
        action_sent = &#34;cancel&#34;
        if action == &#34;send_all&#34; and commands:
            # In remote mode, send the selected commands as a custom block
            # so the server executes exactly what the user picked (e.g., selection &#39;1&#39;)
            action_sent = f&#34;custom:{chr(10).join(commands)}&#34;
        elif action == &#34;custom&#34; and custom_cmd:
            action_sent = f&#34;custom:{chr(10).join(custom_cmd)}&#34;
        request_queue.put(connpy_pb2.InteractRequest(copilot_action=action_sent))
        resume_generator()
        tty.setraw(sys.stdin.fileno())
    @handle_errors
    def connect_node(self, unique_id, sftp=False, debug=False, logger=None):
        import sys
        import select
        import tty
        import termios
        import queue
        import os
        import threading
        request_queue = queue.Queue()
        client_buffer_bytes = bytearray()
        pause_stdin = [False]
        wake_r, wake_w = os.pipe()
        def pause_generator():
            pause_stdin[0] = True
            os.write(wake_w, b&#39;\x00&#39;)
        def resume_generator():
            pause_stdin[0] = False
        def request_generator():
            cols, rows = 80, 24
            try:
@@ -946,8 +1099,25 @@ def set_reserved_names(self, names):
            )
            while True:
-                r, _, _ = select.select([sys.stdin.fileno()], [], [])
+                try:
-                if r:
+                    while True:
                        req = request_queue.get_nowait()
                        if req is None:
                            return
                        yield req
                except queue.Empty:
                    pass
                if pause_stdin[0]:
                    import time
                    time.sleep(0.05)
                    continue
                r, _, _ = select.select([sys.stdin.fileno(), wake_r], [], [], 0.05)
                if wake_r in r:
                    os.read(wake_r, 1)
                    continue
                if sys.stdin.fileno() in r and not pause_stdin[0]:
                    try:
                        data = os.read(sys.stdin.fileno(), 1024)
                        if not data:
@@ -969,30 +1139,74 @@ def set_reserved_names(self, names):
        old_tty = termios.tcgetattr(sys.stdin)
        try:
            import time
            tty.setraw(sys.stdin.fileno())
            response_iterator = self.stub.interact_node(request_generator())
-            # First response is connection status
+            import queue
            response_queue = queue.Queue()
            def response_consumer():
                try:
-                first_res = next(response_iterator)
+                    for r in response_iterator:
-                if first_res.success:
+                        response_queue.put(r)
                except Exception:
                    pass
                response_queue.put(None)
            t_consumer = threading.Thread(target=response_consumer, daemon=True)
            t_consumer.start()
            # First phase: Wait for connection status, print early data
            try:
                while True:
                    res = response_queue.get()
                    if res is None:
                        return
                    if res.stdout_data:
                        data = res.stdout_data
                        if debug:
                            data = data.replace(b&#39;\x1b[H\x1b[2J&#39;, b&#39;&#39;).replace(b&#39;\x1bc&#39;, b&#39;&#39;).replace(b&#39;\x1b[3J&#39;, b&#39;&#39;)
                        os.write(sys.stdout.fileno(), data)
                    if res.success:
                        # Connection established on server, show success message
                        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
                        printer.success(conn_msg)
                        pause_stdin[0] = False
                        tty.setraw(sys.stdin.fileno())
-                else:
+                        break
                    if res.error_message:
                        # Connection failed on server
                        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
-                    printer.error(f&#34;Connection failed: {first_res.error_message}&#34;)
+                        printer.error(f&#34;Connection failed: {res.error_message}&#34;)
                        return
-            except StopIteration:
+            except queue.Empty:
                return
-            for res in response_iterator:
+            # Second phase: Stream active session
            # Clear screen filter is only applied before success (Phase 1).
            # Once the user has a prompt, Ctrl+L must work normally.
            while True:
                res = response_queue.get()
                if res is None:
                    break
                if res.copilot_prompt:
                    self._handle_remote_copilot(
                        res, request_queue, response_queue, 
                        client_buffer_bytes, 
                        pause_generator, resume_generator, old_tty
                    )
                    continue
                if res.stdout_data:
                    os.write(sys.stdout.fileno(), res.stdout_data)
                    client_buffer_bytes.extend(res.stdout_data)
        finally:
            termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
            os.close(wake_r)
            os.close(wake_w)
    @handle_errors
    def connect_dynamic(self, connection_params, debug=False):
@@ -1000,10 +1214,22 @@ def set_reserved_names(self, names):
        import select
        import tty
        import termios
        import queue
        import os
        import json
        params_json = json.dumps(connection_params)
        request_queue = queue.Queue()
        client_buffer_bytes = bytearray()
        pause_stdin = [False]
        wake_r, wake_w = os.pipe()
        def pause_generator():
            pause_stdin[0] = True
            os.write(wake_w, b&#39;\x00&#39;)
        def resume_generator():
            pause_stdin[0] = False
        def request_generator():
            cols, rows = 80, 24
@@ -1019,8 +1245,25 @@ def set_reserved_names(self, names):
            )
            while True:
-                r, _, _ = select.select([sys.stdin.fileno()], [], [])
+                try:
-                if r:
+                    while True:
                        req = request_queue.get_nowait()
                        if req is None:
                            return
                        yield req
                except queue.Empty:
                    pass
                if pause_stdin[0]:
                    import time
                    time.sleep(0.05)
                    continue
                r, _, _ = select.select([sys.stdin.fileno(), wake_r], [], [], 0.05)
                if wake_r in r:
                    os.read(wake_r, 1)
                    continue
                if sys.stdin.fileno() in r and not pause_stdin[0]:
                    try:
                        data = os.read(sys.stdin.fileno(), 1024)
                        if not data:
@@ -1043,30 +1286,72 @@ def set_reserved_names(self, names):
        old_tty = termios.tcgetattr(sys.stdin)
        try:
            import time
            tty.setraw(sys.stdin.fileno())
            response_iterator = self.stub.interact_node(request_generator())
-            # First response is connection status
+            import queue
            response_queue = queue.Queue()
            def response_consumer():
                try:
-                first_res = next(response_iterator)
+                    for r in response_iterator:
-                if first_res.success:
+                        response_queue.put(r)
                except Exception:
                    pass
                response_queue.put(None)
            t_consumer = threading.Thread(target=response_consumer, daemon=True)
            t_consumer.start()
            # First phase: Wait for connection status, print early data
            try:
                while True:
                    res = response_queue.get()
                    if res is None:
                        return
                    if res.stdout_data:
                        data = res.stdout_data
                        if debug:
                            data = data.replace(b&#39;\x1b[H\x1b[2J&#39;, b&#39;&#39;).replace(b&#39;\x1bc&#39;, b&#39;&#39;).replace(b&#39;\x1b[3J&#39;, b&#39;&#39;)
                        os.write(sys.stdout.fileno(), data)
                    if res.success:
                        # Connection established on server, show success message
                        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
                        printer.success(conn_msg)
                        pause_stdin[0] = False
                        tty.setraw(sys.stdin.fileno())
-                else:
+                        break
                    if res.error_message:
                        # Connection failed on server
                        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
-                    printer.error(f&#34;Connection failed: {first_res.error_message}&#34;)
+                        printer.error(f&#34;Connection failed: {res.error_message}&#34;)
                        return
-            except StopIteration:
+            except queue.Empty:
                return
-            for res in response_iterator:
+            # Second phase: Stream active session
            while True:
                res = response_queue.get()
                if res is None:
                    break
                if res.copilot_prompt:
                    self._handle_remote_copilot(
                        res, request_queue, response_queue, 
                        client_buffer_bytes, 
                        pause_generator, resume_generator, old_tty
                    )
                    continue
                if res.stdout_data:
                    os.write(sys.stdout.fileno(), res.stdout_data)
                    client_buffer_bytes.extend(res.stdout_data)
        finally:
            termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
            os.close(wake_r)
            os.close(wake_w)
    @MethodHook
    @handle_errors
@@ -1220,10 +1505,22 @@ def connect_dynamic(self, connection_params, debug=False):
    import select
    import tty
    import termios
    import queue
    import os
    import json
    params_json = json.dumps(connection_params)
    request_queue = queue.Queue()
    client_buffer_bytes = bytearray()
    pause_stdin = [False]
    wake_r, wake_w = os.pipe()
    def pause_generator():
        pause_stdin[0] = True
        os.write(wake_w, b&#39;\x00&#39;)
    def resume_generator():
        pause_stdin[0] = False
    def request_generator():
        cols, rows = 80, 24
@@ -1239,8 +1536,25 @@ def connect_dynamic(self, connection_params, debug=False):
        )
        while True:
-            r, _, _ = select.select([sys.stdin.fileno()], [], [])
+            try:
-            if r:
+                while True:
                    req = request_queue.get_nowait()
                    if req is None:
                        return
                    yield req
            except queue.Empty:
                pass
            if pause_stdin[0]:
                import time
                time.sleep(0.05)
                continue
            r, _, _ = select.select([sys.stdin.fileno(), wake_r], [], [], 0.05)
            if wake_r in r:
                os.read(wake_r, 1)
                continue
            if sys.stdin.fileno() in r and not pause_stdin[0]:
                try:
                    data = os.read(sys.stdin.fileno(), 1024)
                    if not data:
@@ -1263,30 +1577,72 @@ def connect_dynamic(self, connection_params, debug=False):
    old_tty = termios.tcgetattr(sys.stdin)
    try:
        import time
        tty.setraw(sys.stdin.fileno())
        response_iterator = self.stub.interact_node(request_generator())
-        # First response is connection status
+        import queue
        response_queue = queue.Queue()
        def response_consumer():
            try:
-            first_res = next(response_iterator)
+                for r in response_iterator:
-            if first_res.success:
+                    response_queue.put(r)
            except Exception:
                pass
            response_queue.put(None)
        t_consumer = threading.Thread(target=response_consumer, daemon=True)
        t_consumer.start()
        # First phase: Wait for connection status, print early data
        try:
            while True:
                res = response_queue.get()
                if res is None:
                    return
                if res.stdout_data:
                    data = res.stdout_data
                    if debug:
                        data = data.replace(b&#39;\x1b[H\x1b[2J&#39;, b&#39;&#39;).replace(b&#39;\x1bc&#39;, b&#39;&#39;).replace(b&#39;\x1b[3J&#39;, b&#39;&#39;)
                    os.write(sys.stdout.fileno(), data)
                if res.success:
                    # Connection established on server, show success message
                    termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
                    printer.success(conn_msg)
                    pause_stdin[0] = False
                    tty.setraw(sys.stdin.fileno())
-            else:
+                    break
                if res.error_message:
                    # Connection failed on server
                    termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
-                printer.error(f&#34;Connection failed: {first_res.error_message}&#34;)
+                    printer.error(f&#34;Connection failed: {res.error_message}&#34;)
                    return
-        except StopIteration:
+        except queue.Empty:
            return
-        for res in response_iterator:
+        # Second phase: Stream active session
        while True:
            res = response_queue.get()
            if res is None:
                break
            if res.copilot_prompt:
                self._handle_remote_copilot(
                    res, request_queue, response_queue, 
                    client_buffer_bytes, 
                    pause_generator, resume_generator, old_tty
                )
                continue
            if res.stdout_data:
                os.write(sys.stdout.fileno(), res.stdout_data)
                client_buffer_bytes.extend(res.stdout_data)
    finally:
-        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)</code></pre>
+        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
        os.close(wake_r)
        os.close(wake_w)</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
@@ -1304,9 +1660,22 @@ def connect_node(self, unique_id, sftp=False, debug=False, logger=None):
    import select
    import tty
    import termios
    import queue
    import os
    import threading
    request_queue = queue.Queue()
    client_buffer_bytes = bytearray()
    pause_stdin = [False]
    wake_r, wake_w = os.pipe()
    def pause_generator():
        pause_stdin[0] = True
        os.write(wake_w, b&#39;\x00&#39;)
    def resume_generator():
        pause_stdin[0] = False
    def request_generator():
        cols, rows = 80, 24
        try:
@@ -1320,8 +1689,25 @@ def connect_node(self, unique_id, sftp=False, debug=False, logger=None):
        )
        while True:
-            r, _, _ = select.select([sys.stdin.fileno()], [], [])
+            try:
-            if r:
+                while True:
                    req = request_queue.get_nowait()
                    if req is None:
                        return
                    yield req
            except queue.Empty:
                pass
            if pause_stdin[0]:
                import time
                time.sleep(0.05)
                continue
            r, _, _ = select.select([sys.stdin.fileno(), wake_r], [], [], 0.05)
            if wake_r in r:
                os.read(wake_r, 1)
                continue
            if sys.stdin.fileno() in r and not pause_stdin[0]:
                try:
                    data = os.read(sys.stdin.fileno(), 1024)
                    if not data:
@@ -1343,30 +1729,74 @@ def connect_node(self, unique_id, sftp=False, debug=False, logger=None):
    old_tty = termios.tcgetattr(sys.stdin)
    try:
        import time
        tty.setraw(sys.stdin.fileno())
        response_iterator = self.stub.interact_node(request_generator())
-        # First response is connection status
+        import queue
        response_queue = queue.Queue()
        def response_consumer():
            try:
-            first_res = next(response_iterator)
+                for r in response_iterator:
-            if first_res.success:
+                    response_queue.put(r)
            except Exception:
                pass
            response_queue.put(None)
        t_consumer = threading.Thread(target=response_consumer, daemon=True)
        t_consumer.start()
        # First phase: Wait for connection status, print early data
        try:
            while True:
                res = response_queue.get()
                if res is None:
                    return
                if res.stdout_data:
                    data = res.stdout_data
                    if debug:
                        data = data.replace(b&#39;\x1b[H\x1b[2J&#39;, b&#39;&#39;).replace(b&#39;\x1bc&#39;, b&#39;&#39;).replace(b&#39;\x1b[3J&#39;, b&#39;&#39;)
                    os.write(sys.stdout.fileno(), data)
                if res.success:
                    # Connection established on server, show success message
                    termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
                    printer.success(conn_msg)
                    pause_stdin[0] = False
                    tty.setraw(sys.stdin.fileno())
-            else:
+                    break
                if res.error_message:
                    # Connection failed on server
                    termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
-                printer.error(f&#34;Connection failed: {first_res.error_message}&#34;)
+                    printer.error(f&#34;Connection failed: {res.error_message}&#34;)
                    return
-        except StopIteration:
+        except queue.Empty:
            return
-        for res in response_iterator:
+        # Second phase: Stream active session
        # Clear screen filter is only applied before success (Phase 1).
        # Once the user has a prompt, Ctrl+L must work normally.
        while True:
            res = response_queue.get()
            if res is None:
                break
            if res.copilot_prompt:
                self._handle_remote_copilot(
                    res, request_queue, response_queue, 
                    client_buffer_bytes, 
                    pause_generator, resume_generator, old_tty
                )
                continue
            if res.stdout_data:
                os.write(sys.stdout.fileno(), res.stdout_data)
                client_buffer_bytes.extend(res.stdout_data)
    finally:
-        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)</code></pre>
+        termios.tcsetattr(sys.stdin, termios.TCSADRAIN, old_tty)
        os.close(wake_r)
        os.close(wake_w)</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
@@ -2036,6 +2466,7 @@ def stop_api(self):
 <h4><code><a title="connpy.grpc_layer.stubs.AIStub" href="#connpy.grpc_layer.stubs.AIStub">AIStub</a></code></h4>
 <ul class="two-column">
 <li><code><a title="connpy.grpc_layer.stubs.AIStub.ask" href="#connpy.grpc_layer.stubs.AIStub.ask">ask</a></code></li>
 <li><code><a title="connpy.grpc_layer.stubs.AIStub.configure_mcp" href="#connpy.grpc_layer.stubs.AIStub.configure_mcp">configure_mcp</a></code></li>
 <li><code><a title="connpy.grpc_layer.stubs.AIStub.configure_provider" href="#connpy.grpc_layer.stubs.AIStub.configure_provider">configure_provider</a></code></li>
 <li><code><a title="connpy.grpc_layer.stubs.AIStub.confirm" href="#connpy.grpc_layer.stubs.AIStub.confirm">confirm</a></code></li>
 <li><code><a title="connpy.grpc_layer.stubs.AIStub.delete_session" href="#connpy.grpc_layer.stubs.AIStub.delete_session">delete_session</a></code></li>
@@ -2130,7 +2561,7 @@ def stop_api(self):
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.grpc_layer.utils API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -138,7 +138,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -0,0 +1,349 @@
 <!doctype html>
 <html lang="en">
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
 <meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.mcp_client API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/typography.min.css" integrity="sha512-Y1DYSb995BAfxobCkKepB1BqJJTPrOp3zPL74AWFugHHmmdcvO+C48WLrUOlhGMc0QG7AE3f7gmvvcrmX2fDoA==" crossorigin>
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/styles/default.min.css" crossorigin>
 <style>:root{--highlight-color:#fe9}.flex{display:flex !important}body{line-height:1.5em}#content{padding:20px}#sidebar{padding:1.5em;overflow:hidden}#sidebar > *:last-child{margin-bottom:2cm}.http-server-breadcrumbs{font-size:130%;margin:0 0 15px 0}#footer{font-size:.75em;padding:5px 30px;border-top:1px solid #ddd;text-align:right}#footer p{margin:0 0 0 1em;display:inline-block}#footer p:last-child{margin-right:30px}h1,h2,h3,h4,h5{font-weight:300}h1{font-size:2.5em;line-height:1.1em}h2{font-size:1.75em;margin:2em 0 .50em 0}h3{font-size:1.4em;margin:1.6em 0 .7em 0}h4{margin:0;font-size:105%}h1:target,h2:target,h3:target,h4:target,h5:target,h6:target{background:var(--highlight-color);padding:.2em 0}a{color:#058;text-decoration:none;transition:color .2s ease-in-out}a:visited{color:#503}a:hover{color:#b62}.title code{font-weight:bold}h2[id^="header-"]{margin-top:2em}.ident{color:#900;font-weight:bold}pre code{font-size:.8em;line-height:1.4em;padding:1em;display:block}code{background:#f3f3f3;font-family:"DejaVu Sans Mono",monospace;padding:1px 4px;overflow-wrap:break-word}h1 code{background:transparent}pre{border-top:1px solid #ccc;border-bottom:1px solid #ccc;margin:1em 0}#http-server-module-list{display:flex;flex-flow:column}#http-server-module-list div{display:flex}#http-server-module-list dt{min-width:10%}#http-server-module-list p{margin-top:0}.toc ul,#index{list-style-type:none;margin:0;padding:0}#index code{background:transparent}#index h3{border-bottom:1px solid #ddd}#index ul{padding:0}#index h4{margin-top:.6em;font-weight:bold}@media (min-width:200ex){#index .two-column{column-count:2}}@media (min-width:300ex){#index .two-column{column-count:3}}dl{margin-bottom:2em}dl dl:last-child{margin-bottom:4em}dd{margin:0 0 1em 3em}#header-classes + dl > dd{margin-bottom:3em}dd dd{margin-left:2em}dd p{margin:10px 0}.name{background:#eee;font-size:.85em;padding:5px 10px;display:inline-block;min-width:40%}.name:hover{background:#e0e0e0}dt:target .name{background:var(--highlight-color)}.name > span:first-child{white-space:nowrap}.name.class > span:nth-child(2){margin-left:.4em}.inherited{color:#999;border-left:5px solid #eee;padding-left:1em}.inheritance em{font-style:normal;font-weight:bold}.desc h2{font-weight:400;font-size:1.25em}.desc h3{font-size:1em}.desc dt code{background:inherit}.source > summary,.git-link-div{color:#666;text-align:right;font-weight:400;font-size:.8em;text-transform:uppercase}.source summary > *{white-space:nowrap;cursor:pointer}.git-link{color:inherit;margin-left:1em}.source pre{max-height:500px;overflow:auto;margin:0}.source pre code{font-size:12px;overflow:visible;min-width:max-content}.hlist{list-style:none}.hlist li{display:inline}.hlist li:after{content:',\2002'}.hlist li:last-child:after{content:none}.hlist .hlist{display:inline;padding-left:1em}img{max-width:100%}td{padding:0 .5em}.admonition{padding:.1em 1em;margin:1em 0}.admonition-title{font-weight:bold}.admonition.note,.admonition.info,.admonition.important{background:#aef}.admonition.todo,.admonition.versionadded,.admonition.tip,.admonition.hint{background:#dfd}.admonition.warning,.admonition.versionchanged,.admonition.deprecated{background:#fd4}.admonition.error,.admonition.danger,.admonition.caution{background:lightpink}</style>
 <style media="screen and (min-width: 700px)">@media screen and (min-width:700px){#sidebar{width:30%;height:100vh;overflow:auto;position:sticky;top:0}#content{width:70%;max-width:100ch;padding:3em 4em;border-left:1px solid #ddd}pre code{font-size:1em}.name{font-size:1em}main{display:flex;flex-direction:row-reverse;justify-content:flex-end}.toc ul ul,#index ul ul{padding-left:1em}.toc > ul > li{margin-top:.5em}}</style>
 <style media="print">@media print{#sidebar h1{page-break-before:always}.source{display:none}}@media print{*{background:transparent !important;color:#000 !important;box-shadow:none !important;text-shadow:none !important}a[href]:after{content:" (" attr(href) ")";font-size:90%}a[href][title]:after{content:none}abbr[title]:after{content:" (" attr(title) ")"}.ir a:after,a[href^="javascript:"]:after,a[href^="#"]:after{content:""}pre,blockquote{border:1px solid #999;page-break-inside:avoid}thead{display:table-header-group}tr,img{page-break-inside:avoid}img{max-width:100% !important}@page{margin:0.5cm}p,h2,h3{orphans:3;widows:3}h1,h2,h3,h4,h5,h6{page-break-after:avoid}}</style>
 <script defer src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/highlight.min.js" integrity="sha512-D9gUyxqja7hBtkWpPWGt9wfbfaMGVt9gnyCvYa+jojwwPHLCzUm5i8rpk7vD7wNee9bA35eYIjobYPaQuKS1MQ==" crossorigin></script>
 <script>window.addEventListener('DOMContentLoaded', () => {
 hljs.configure({languages: ['bash', 'css', 'diff', 'graphql', 'ini', 'javascript', 'json', 'plaintext', 'python', 'python-repl', 'rust', 'shell', 'sql', 'typescript', 'xml', 'yaml']});
 hljs.highlightAll();
 /* Collapse source docstrings */
 setTimeout(() => {
 [...document.querySelectorAll('.hljs.language-python > .hljs-string')]
 .filter(el => el.innerHTML.length > 200 && ['"""', "'''"].includes(el.innerHTML.substring(0, 3)))
 .forEach(el => {
 let d = document.createElement('details');
 d.classList.add('hljs-string');
 d.innerHTML = '<summary>"""</summary>' + el.innerHTML.substring(3);
 el.replaceWith(d);
 });
 }, 100);
 })</script>
 </head>
 <body>
 <main>
 <article id="content">
 <header>
 <h1 class="title">Module <code>connpy.mcp_client</code></h1>
 </header>
 <section id="section-intro">
 </section>
 <section>
 </section>
 <section>
 </section>
 <section>
 </section>
 <section>
 <h2 class="section-title" id="header-classes">Classes</h2>
 <dl>
 <dt id="connpy.mcp_client.MCPClientManager"><code class="flex name class">
 <span>class <span class="ident">MCPClientManager</span></span>
 <span>(</span><span>config=None)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">class MCPClientManager:
    &#34;&#34;&#34;Manages MCP SSE client connections for connpy.&#34;&#34;&#34;
    _instance = None
    _lock = threading.Lock()
    def __new__(cls, *args, **kwargs):
        with cls._lock:
            if cls._instance is None:
                cls._instance = super(MCPClientManager, cls).__new__(cls)
                cls._instance._initialized = False
            return cls._instance
    def __init__(self, config=None):
        if self._initialized:
            return
        self.config = config
        self.sessions: Dict[str, Dict[str, Any]] = {} # name -&gt; {session, stack}
        self.tool_cache: Dict[str, List[Dict[str, Any]]] = {}
        self._connecting: Dict[str, asyncio.Future] = {}
        self._initialized = True
    async def get_tools_for_llm(self, os_filter: Optional[str] = None) -&gt; List[Dict[str, Any]]:
        &#34;&#34;&#34;
        Fetches tools from enabled MCP servers that match the OS filter.
        &#34;&#34;&#34;
        if not MCP_AVAILABLE:
            return []
        all_llm_tools = []
        try:
            mcp_config = self.config.config.get(&#34;ai&#34;, {}).get(&#34;mcp_servers&#34;, {})
        except Exception:
            return []
        async def _fetch(name, cfg):
            if not cfg.get(&#34;enabled&#34;, True): return []
            # Filter by OS if specified in config (primarily used for copilot strict matching)
            auto_os = cfg.get(&#34;auto_load_on_os&#34;)
            if os_filter is not None and auto_os and os_filter.lower() != auto_os.lower():
                return []
            try:
                session = await self._ensure_connected(name, cfg)
                if session:
                    if name in self.tool_cache: return self.tool_cache[name]
                    llm_tools = await self._fetch_tools_as_openai(name, session)
                    self.tool_cache[name] = llm_tools
                    return llm_tools
            except Exception:
                pass
            return []
        tasks = [ _fetch(name, cfg) for name, cfg in mcp_config.items() ]
        if tasks:
            results = await asyncio.gather(*tasks)
            for tools in results:
                all_llm_tools.extend(tools)
        return all_llm_tools
    async def _ensure_connected(self, name: str, cfg: Dict[str, Any]) -&gt; Optional[Any]:
        if not MCP_AVAILABLE: return None
        if name in self.sessions and self.sessions[name].get(&#34;session&#34;):
            return self.sessions[name][&#34;session&#34;]
        url = cfg.get(&#34;url&#34;)
        if not url:
            return None
        if name in self._connecting:
            try:
                return await asyncio.wait_for(asyncio.shield(self._connecting[name]), timeout=10.0)
            except Exception:
                return None
        loop = asyncio.get_running_loop()
        fut = loop.create_future()
        self._connecting[name] = fut
        try:
            from contextlib import AsyncExitStack
            stack = AsyncExitStack()
            async def _do_connect():
                read, write = await stack.enter_async_context(sse_client(url))
                session = await stack.enter_async_context(ClientSession(read, write))
                await session.initialize()
                return session
            session = await asyncio.wait_for(_do_connect(), timeout=15.0)
            self.sessions[name] = {&#34;session&#34;: session, &#34;stack&#34;: stack}
            fut.set_result(session)
            return session
        except Exception:
            fut.set_result(None)
            return None
        finally:
            if name in self._connecting:
                del self._connecting[name]
    async def _fetch_tools_as_openai(self, server_name: str, session: Any) -&gt; List[Dict[str, Any]]:
        try:
            result = await asyncio.wait_for(session.list_tools(), timeout=5.0)
            openai_tools = []
            for tool in result.tools:
                # Use mcp_ prefix to ensure valid function name for LiteLLM/Gemini
                prefixed_name = f&#34;mcp_{server_name}__{tool.name}&#34;
                openai_tools.append({
                    &#34;type&#34;: &#34;function&#34;,
                    &#34;function&#34;: {
                        &#34;name&#34;: prefixed_name,
                        &#34;description&#34;: f&#34;[{server_name}] {tool.description}&#34;,
                        &#34;parameters&#34;: tool.inputSchema
                    }
                })
            return openai_tools
        except Exception:
            return []
    async def call_tool(self, full_tool_name: str, arguments: Dict[str, Any]) -&gt; Any:
        &#34;&#34;&#34;Calls an MCP tool and returns text result.&#34;&#34;&#34;
        if not MCP_AVAILABLE:
            return &#34;Error: MCP SDK is not installed.&#34;
        if &#34;__&#34; not in full_tool_name:
            return f&#34;Error: Tool {full_tool_name} is not a valid MCP tool.&#34;
        clean_name = full_tool_name[4:] if full_tool_name.startswith(&#34;mcp_&#34;) else full_tool_name
        server_name, tool_name = clean_name.split(&#34;__&#34;, 1)
        if server_name not in self.sessions:
            return f&#34;Error: MCP server {server_name} is not connected.&#34;
        session = self.sessions[server_name][&#34;session&#34;]
        try:
            result = await asyncio.wait_for(session.call_tool(tool_name, arguments), timeout=60.0)
            text_outputs = [content.text for content in result.content if hasattr(content, &#34;text&#34;)]
            return &#34;\n&#34;.join(text_outputs) if text_outputs else str(result)
        except Exception as e:
            return f&#34;Error calling tool {tool_name} on {server_name}: {str(e)}&#34;
    async def shutdown(self):
        &#34;&#34;&#34;Close all SSE connections.&#34;&#34;&#34;
        for name, data in self.sessions.items():
            stack = data.get(&#34;stack&#34;)
            if stack:
                await stack.aclose()
        self.sessions = {}</code></pre>
 </details>
 <div class="desc"><p>Manages MCP SSE client connections for connpy.</p></div>
 <h3>Methods</h3>
 <dl>
 <dt id="connpy.mcp_client.MCPClientManager.call_tool"><code class="name flex">
 <span>async def <span class="ident">call_tool</span></span>(<span>self, full_tool_name: str, arguments: Dict[str, Any]) ‑> Any</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">async def call_tool(self, full_tool_name: str, arguments: Dict[str, Any]) -&gt; Any:
    &#34;&#34;&#34;Calls an MCP tool and returns text result.&#34;&#34;&#34;
    if not MCP_AVAILABLE:
        return &#34;Error: MCP SDK is not installed.&#34;
    if &#34;__&#34; not in full_tool_name:
        return f&#34;Error: Tool {full_tool_name} is not a valid MCP tool.&#34;
    clean_name = full_tool_name[4:] if full_tool_name.startswith(&#34;mcp_&#34;) else full_tool_name
    server_name, tool_name = clean_name.split(&#34;__&#34;, 1)
    if server_name not in self.sessions:
        return f&#34;Error: MCP server {server_name} is not connected.&#34;
    session = self.sessions[server_name][&#34;session&#34;]
    try:
        result = await asyncio.wait_for(session.call_tool(tool_name, arguments), timeout=60.0)
        text_outputs = [content.text for content in result.content if hasattr(content, &#34;text&#34;)]
        return &#34;\n&#34;.join(text_outputs) if text_outputs else str(result)
    except Exception as e:
        return f&#34;Error calling tool {tool_name} on {server_name}: {str(e)}&#34;</code></pre>
 </details>
 <div class="desc"><p>Calls an MCP tool and returns text result.</p></div>
 </dd>
 <dt id="connpy.mcp_client.MCPClientManager.get_tools_for_llm"><code class="name flex">
 <span>async def <span class="ident">get_tools_for_llm</span></span>(<span>self, os_filter: str | None = None) ‑> List[Dict[str, Any]]</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">async def get_tools_for_llm(self, os_filter: Optional[str] = None) -&gt; List[Dict[str, Any]]:
    &#34;&#34;&#34;
    Fetches tools from enabled MCP servers that match the OS filter.
    &#34;&#34;&#34;
    if not MCP_AVAILABLE:
        return []
    all_llm_tools = []
    try:
        mcp_config = self.config.config.get(&#34;ai&#34;, {}).get(&#34;mcp_servers&#34;, {})
    except Exception:
        return []
    async def _fetch(name, cfg):
        if not cfg.get(&#34;enabled&#34;, True): return []
        # Filter by OS if specified in config (primarily used for copilot strict matching)
        auto_os = cfg.get(&#34;auto_load_on_os&#34;)
        if os_filter is not None and auto_os and os_filter.lower() != auto_os.lower():
            return []
        try:
            session = await self._ensure_connected(name, cfg)
            if session:
                if name in self.tool_cache: return self.tool_cache[name]
                llm_tools = await self._fetch_tools_as_openai(name, session)
                self.tool_cache[name] = llm_tools
                return llm_tools
        except Exception:
            pass
        return []
    tasks = [ _fetch(name, cfg) for name, cfg in mcp_config.items() ]
    if tasks:
        results = await asyncio.gather(*tasks)
        for tools in results:
            all_llm_tools.extend(tools)
    return all_llm_tools</code></pre>
 </details>
 <div class="desc"><p>Fetches tools from enabled MCP servers that match the OS filter.</p></div>
 </dd>
 <dt id="connpy.mcp_client.MCPClientManager.shutdown"><code class="name flex">
 <span>async def <span class="ident">shutdown</span></span>(<span>self)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">async def shutdown(self):
    &#34;&#34;&#34;Close all SSE connections.&#34;&#34;&#34;
    for name, data in self.sessions.items():
        stack = data.get(&#34;stack&#34;)
        if stack:
            await stack.aclose()
    self.sessions = {}</code></pre>
 </details>
 <div class="desc"><p>Close all SSE connections.</p></div>
 </dd>
 </dl>
 </dd>
 </dl>
 </section>
 </article>
 <nav id="sidebar">
 <div class="toc">
 <ul></ul>
 </div>
 <ul id="index">
 <li><h3>Super-module</h3>
 <ul>
 <li><code><a title="connpy" href="index.html">connpy</a></code></li>
 </ul>
 </li>
 <li><h3><a href="#header-classes">Classes</a></h3>
 <ul>
 <li>
 <h4><code><a title="connpy.mcp_client.MCPClientManager" href="#connpy.mcp_client.MCPClientManager">MCPClientManager</a></code></h4>
 <ul class="">
 <li><code><a title="connpy.mcp_client.MCPClientManager.call_tool" href="#connpy.mcp_client.MCPClientManager.call_tool">call_tool</a></code></li>
 <li><code><a title="connpy.mcp_client.MCPClientManager.get_tools_for_llm" href="#connpy.mcp_client.MCPClientManager.get_tools_for_llm">get_tools_for_llm</a></code></li>
 <li><code><a title="connpy.mcp_client.MCPClientManager.shutdown" href="#connpy.mcp_client.MCPClientManager.shutdown">shutdown</a></code></li>
 </ul>
 </li>
 </ul>
 </li>
 </ul>
 </nav>
 </main>
 <footer id="footer">
 <p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.proto API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -60,7 +60,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.ai_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -58,6 +58,138 @@ el.replaceWith(d);
 <pre><code class="python">class AIService(BaseService):
    &#34;&#34;&#34;Business logic for interacting with AI agents and LLM configurations.&#34;&#34;&#34;
    def build_context_blocks(self, raw_bytes: bytes, cmd_byte_positions: list, node_info: dict, last_line: str = &#34;&#34;) -&gt; list:
        &#34;&#34;&#34;Identifies command blocks in the terminal history.&#34;&#34;&#34;
        blocks = []
        if not raw_bytes:
            return blocks
        default_prompt = r&#39;&gt;$|#$|\$$|&gt;.$|#.$|\$.$&#39;
        device_prompt = node_info.get(&#34;prompt&#34;, default_prompt) if isinstance(node_info, dict) else default_prompt
        prompt_re_str = re.sub(r&#39;(?&lt;!\\)\$&#39;, &#39;&#39;, device_prompt)
        try:
            prompt_re = re.compile(prompt_re_str)
        except Exception:
            prompt_re = re.compile(re.sub(r&#39;(?&lt;!\\)\$&#39;, &#39;&#39;, default_prompt))
        parsed_positions = []
        if cmd_byte_positions and len(cmd_byte_positions) &gt;= 1:
            for i in range(1, len(cmd_byte_positions)):
                pos, known_cmd = cmd_byte_positions[i]
                prev_pos = cmd_byte_positions[i-1][0]
                if known_cmd:
                    prev_chunk = raw_bytes[prev_pos:pos]
                    prev_cleaned = log_cleaner(prev_chunk.decode(errors=&#39;replace&#39;))
                    prev_lines = [l for l in prev_cleaned.split(&#39;\n&#39;) if l.strip()]
                    prompt_text = prev_lines[-1].strip() if prev_lines else &#34;&#34;
                    preview = f&#34;{prompt_text}{known_cmd}&#34; if prompt_text else known_cmd
                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
                else:
                    chunk = raw_bytes[prev_pos:pos]
                    cleaned = log_cleaner(chunk.decode(errors=&#39;replace&#39;))
                    lines = [l for l in cleaned.split(&#39;\n&#39;) if l.strip()]
                    preview = lines[-1].strip() if lines else &#34;&#34;
                    if preview:
                        match = prompt_re.search(preview)
                        if match:
                            cmd_text = preview[match.end():].strip()
                            if cmd_text:
                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
                            else:
                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;EMPTY_PROMPT&#34;, &#34;preview&#34;: &#34;&#34;})
                        else:
                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
                    else:
                        parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
        last_newline = raw_bytes.rfind(b&#39;\n&#39;)
        current_prompt_pos = last_newline + 1 if last_newline != -1 else 0
        current_end = len(raw_bytes)
        for i, item in enumerate(parsed_positions):
            if item[&#34;type&#34;] == &#34;VALID_CMD&#34;:
                start_pos = item[&#34;pos&#34;]
                preview = item[&#34;preview&#34;]
                # Find the end position: next VALID_CMD or EMPTY_PROMPT
                end_pos = current_prompt_pos
                for j in range(i + 1, len(parsed_positions)):
                    next_item = parsed_positions[j]
                    if next_item[&#34;type&#34;] in (&#34;VALID_CMD&#34;, &#34;EMPTY_PROMPT&#34;):
                        end_pos = next_item[&#34;pos&#34;]
                        break
                blocks.append((start_pos, end_pos, preview))
        # Always ensure there is a final block representing the current prompt
        if not blocks:
            blocks.append((current_prompt_pos, current_end, last_line[:80] if last_line else &#34;CURRENT CONTEXT&#34;))
        elif blocks[-1][0] &lt; current_prompt_pos:
            blocks.append((current_prompt_pos, current_end, last_line[:80] if last_line else &#34;CURRENT CONTEXT&#34;))
        return blocks
    def process_copilot_input(self, input_text: str, session_state: dict) -&gt; dict:
        &#34;&#34;&#34;Parses slash commands and manages session state. Returns directive dict.&#34;&#34;&#34;
        text = input_text.strip()
        if not text.startswith(&#39;/&#39;):
            return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: text, &#34;overrides&#34;: {}}
        parts = text.split(maxsplit=1)
        cmd = parts[0].lower()
        args = parts[1] if len(parts) &gt; 1 else &#34;&#34;
        # 1. State Commands (Persistent)
        if cmd == &#34;/os&#34;:
            if args:
                session_state[&#39;os&#39;] = args
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: f&#34;OS context changed to {args}&#34;}
        elif cmd == &#34;/prompt&#34;:
            if args:
                session_state[&#39;prompt&#39;] = args
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: f&#34;Prompt regex changed to {args}&#34;}
        elif cmd == &#34;/memorize&#34;:
            if args:
                session_state[&#39;memories&#39;].append(args)
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: f&#34;Memory added: {args}&#34;}
        elif cmd == &#34;/clear&#34;:
            session_state[&#39;memories&#39;] = []
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Memory cleared&#34;}
        # 2. Hybrid Commands
        elif cmd == &#34;/architect&#34;:
            if not args:
                session_state[&#39;persona&#39;] = &#39;architect&#39;
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Persona set to Architect&#34;}
            else:
                return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;persona&#34;: &#34;architect&#34;}}
        elif cmd == &#34;/engineer&#34;:
            if not args:
                session_state[&#39;persona&#39;] = &#39;engineer&#39;
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Persona set to Engineer&#34;}
            else:
                return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;persona&#34;: &#34;engineer&#34;}}
        elif cmd == &#34;/trust&#34;:
            if not args:
                session_state[&#39;trust_mode&#39;] = True
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Auto-execute (trust) enabled for session&#34;}
            else:
                return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;trust&#34;: True}}
        elif cmd == &#34;/untrust&#34;:
            if not args:
                session_state[&#39;trust_mode&#39;] = False
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Auto-execute (trust) disabled for session&#34;}
            else:
                return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;trust&#34;: False}}
        # Unknown command, execute normally
        return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: text, &#34;overrides&#34;: {}}
    def ask(self, input_text, dryrun=False, chat_history=None, status=None, debug=False, session_id=None, console=None, chunk_callback=None, confirm_handler=None, trust=False, **overrides):
        &#34;&#34;&#34;Send a prompt to the AI agent.&#34;&#34;&#34;
        from connpy.ai import ai
@@ -71,6 +203,21 @@ el.replaceWith(d);
        agent = ai(self.config, console=console)
        return agent.confirm(input_text)
    def ask_copilot(self, terminal_buffer, user_question, node_info=None, chunk_callback=None):
        &#34;&#34;&#34;Ask the AI copilot for terminal assistance.&#34;&#34;&#34;
        from connpy.ai import ai, run_ai_async
        agent = ai(self.config)
        future = run_ai_async(agent.aask_copilot(terminal_buffer, user_question, node_info, chunk_callback=chunk_callback))
        return future.result()
    async def aask_copilot(self, terminal_buffer, user_question, node_info=None, chunk_callback=None):
        &#34;&#34;&#34;Ask the AI copilot for terminal assistance asynchronously.&#34;&#34;&#34;
        from connpy.ai import ai, run_ai_async
        import asyncio
        agent = ai(self.config)
        future = run_ai_async(agent.aask_copilot(terminal_buffer, user_question, node_info, chunk_callback=chunk_callback))
        return await asyncio.wrap_future(future)
    def list_sessions(self):
        &#34;&#34;&#34;Return a list of all saved AI sessions.&#34;&#34;&#34;
@@ -99,6 +246,40 @@ el.replaceWith(d);
        self.config.config[&#34;ai&#34;] = settings
        self.config._saveconfig(self.config.file)
    def configure_mcp(self, name, url=None, enabled=None, auto_load_on_os=None, remove=False):
        &#34;&#34;&#34;Update MCP server settings in the configuration with smart merging.&#34;&#34;&#34;
        ai_settings = self.config.config.get(&#34;ai&#34;, {})
        mcp_servers = ai_settings.get(&#34;mcp_servers&#34;, {})
        if remove:
            if name in mcp_servers:
                del mcp_servers[name]
        else:
            # Get existing or new
            server_cfg = mcp_servers.get(name, {})
            # Partial updates
            if url is not None:
                server_cfg[&#34;url&#34;] = url
            if enabled is not None:
                server_cfg[&#34;enabled&#34;] = bool(enabled)
            elif &#34;enabled&#34; not in server_cfg:
                server_cfg[&#34;enabled&#34;] = True # Default for new entries
            if auto_load_on_os is not None:
                if auto_load_on_os == &#34;&#34;: # Explicit clear
                    if &#34;auto_load_on_os&#34; in server_cfg:
                        del server_cfg[&#34;auto_load_on_os&#34;]
                else:
                    server_cfg[&#34;auto_load_on_os&#34;] = auto_load_on_os
            mcp_servers[name] = server_cfg
        ai_settings[&#34;mcp_servers&#34;] = mcp_servers
        self.config.config[&#34;ai&#34;] = ai_settings
        self.config._saveconfig(self.config.file)
    def load_session_data(self, session_id):
        &#34;&#34;&#34;Load a session&#39;s raw data by ID.&#34;&#34;&#34;
        from connpy.ai import ai
@@ -118,6 +299,24 @@ el.replaceWith(d);
 </ul>
 <h3>Methods</h3>
 <dl>
 <dt id="connpy.services.ai_service.AIService.aask_copilot"><code class="name flex">
 <span>async def <span class="ident">aask_copilot</span></span>(<span>self, terminal_buffer, user_question, node_info=None, chunk_callback=None)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">async def aask_copilot(self, terminal_buffer, user_question, node_info=None, chunk_callback=None):
    &#34;&#34;&#34;Ask the AI copilot for terminal assistance asynchronously.&#34;&#34;&#34;
    from connpy.ai import ai, run_ai_async
    import asyncio
    agent = ai(self.config)
    future = run_ai_async(agent.aask_copilot(terminal_buffer, user_question, node_info, chunk_callback=chunk_callback))
    return await asyncio.wrap_future(future)</code></pre>
 </details>
 <div class="desc"><p>Ask the AI copilot for terminal assistance asynchronously.</p></div>
 </dd>
 <dt id="connpy.services.ai_service.AIService.ask"><code class="name flex">
 <span>def <span class="ident">ask</span></span>(<span>self,<br>input_text,<br>dryrun=False,<br>chat_history=None,<br>status=None,<br>debug=False,<br>session_id=None,<br>console=None,<br>chunk_callback=None,<br>confirm_handler=None,<br>trust=False,<br>**overrides)</span>
 </code></dt>
@@ -134,6 +333,150 @@ el.replaceWith(d);
 </details>
 <div class="desc"><p>Send a prompt to the AI agent.</p></div>
 </dd>
 <dt id="connpy.services.ai_service.AIService.ask_copilot"><code class="name flex">
 <span>def <span class="ident">ask_copilot</span></span>(<span>self, terminal_buffer, user_question, node_info=None, chunk_callback=None)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def ask_copilot(self, terminal_buffer, user_question, node_info=None, chunk_callback=None):
    &#34;&#34;&#34;Ask the AI copilot for terminal assistance.&#34;&#34;&#34;
    from connpy.ai import ai, run_ai_async
    agent = ai(self.config)
    future = run_ai_async(agent.aask_copilot(terminal_buffer, user_question, node_info, chunk_callback=chunk_callback))
    return future.result()</code></pre>
 </details>
 <div class="desc"><p>Ask the AI copilot for terminal assistance.</p></div>
 </dd>
 <dt id="connpy.services.ai_service.AIService.build_context_blocks"><code class="name flex">
 <span>def <span class="ident">build_context_blocks</span></span>(<span>self,<br>raw_bytes: bytes,<br>cmd_byte_positions: list,<br>node_info: dict,<br>last_line: str = '') ‑> list</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def build_context_blocks(self, raw_bytes: bytes, cmd_byte_positions: list, node_info: dict, last_line: str = &#34;&#34;) -&gt; list:
    &#34;&#34;&#34;Identifies command blocks in the terminal history.&#34;&#34;&#34;
    blocks = []
    if not raw_bytes:
        return blocks
    default_prompt = r&#39;&gt;$|#$|\$$|&gt;.$|#.$|\$.$&#39;
    device_prompt = node_info.get(&#34;prompt&#34;, default_prompt) if isinstance(node_info, dict) else default_prompt
    prompt_re_str = re.sub(r&#39;(?&lt;!\\)\$&#39;, &#39;&#39;, device_prompt)
    try:
        prompt_re = re.compile(prompt_re_str)
    except Exception:
        prompt_re = re.compile(re.sub(r&#39;(?&lt;!\\)\$&#39;, &#39;&#39;, default_prompt))
    parsed_positions = []
    if cmd_byte_positions and len(cmd_byte_positions) &gt;= 1:
        for i in range(1, len(cmd_byte_positions)):
            pos, known_cmd = cmd_byte_positions[i]
            prev_pos = cmd_byte_positions[i-1][0]
            if known_cmd:
                prev_chunk = raw_bytes[prev_pos:pos]
                prev_cleaned = log_cleaner(prev_chunk.decode(errors=&#39;replace&#39;))
                prev_lines = [l for l in prev_cleaned.split(&#39;\n&#39;) if l.strip()]
                prompt_text = prev_lines[-1].strip() if prev_lines else &#34;&#34;
                preview = f&#34;{prompt_text}{known_cmd}&#34; if prompt_text else known_cmd
                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
            else:
                chunk = raw_bytes[prev_pos:pos]
                cleaned = log_cleaner(chunk.decode(errors=&#39;replace&#39;))
                lines = [l for l in cleaned.split(&#39;\n&#39;) if l.strip()]
                preview = lines[-1].strip() if lines else &#34;&#34;
                if preview:
                    match = prompt_re.search(preview)
                    if match:
                        cmd_text = preview[match.end():].strip()
                        if cmd_text:
                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
                        else:
                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;EMPTY_PROMPT&#34;, &#34;preview&#34;: &#34;&#34;})
                    else:
                        parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
                else:
                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
    last_newline = raw_bytes.rfind(b&#39;\n&#39;)
    current_prompt_pos = last_newline + 1 if last_newline != -1 else 0
    current_end = len(raw_bytes)
    for i, item in enumerate(parsed_positions):
        if item[&#34;type&#34;] == &#34;VALID_CMD&#34;:
            start_pos = item[&#34;pos&#34;]
            preview = item[&#34;preview&#34;]
            # Find the end position: next VALID_CMD or EMPTY_PROMPT
            end_pos = current_prompt_pos
            for j in range(i + 1, len(parsed_positions)):
                next_item = parsed_positions[j]
                if next_item[&#34;type&#34;] in (&#34;VALID_CMD&#34;, &#34;EMPTY_PROMPT&#34;):
                    end_pos = next_item[&#34;pos&#34;]
                    break
            blocks.append((start_pos, end_pos, preview))
    # Always ensure there is a final block representing the current prompt
    if not blocks:
        blocks.append((current_prompt_pos, current_end, last_line[:80] if last_line else &#34;CURRENT CONTEXT&#34;))
    elif blocks[-1][0] &lt; current_prompt_pos:
        blocks.append((current_prompt_pos, current_end, last_line[:80] if last_line else &#34;CURRENT CONTEXT&#34;))
    return blocks</code></pre>
 </details>
 <div class="desc"><p>Identifies command blocks in the terminal history.</p></div>
 </dd>
 <dt id="connpy.services.ai_service.AIService.configure_mcp"><code class="name flex">
 <span>def <span class="ident">configure_mcp</span></span>(<span>self, name, url=None, enabled=None, auto_load_on_os=None, remove=False)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def configure_mcp(self, name, url=None, enabled=None, auto_load_on_os=None, remove=False):
    &#34;&#34;&#34;Update MCP server settings in the configuration with smart merging.&#34;&#34;&#34;
    ai_settings = self.config.config.get(&#34;ai&#34;, {})
    mcp_servers = ai_settings.get(&#34;mcp_servers&#34;, {})
    if remove:
        if name in mcp_servers:
            del mcp_servers[name]
    else:
        # Get existing or new
        server_cfg = mcp_servers.get(name, {})
        # Partial updates
        if url is not None:
            server_cfg[&#34;url&#34;] = url
        if enabled is not None:
            server_cfg[&#34;enabled&#34;] = bool(enabled)
        elif &#34;enabled&#34; not in server_cfg:
            server_cfg[&#34;enabled&#34;] = True # Default for new entries
        if auto_load_on_os is not None:
            if auto_load_on_os == &#34;&#34;: # Explicit clear
                if &#34;auto_load_on_os&#34; in server_cfg:
                    del server_cfg[&#34;auto_load_on_os&#34;]
            else:
                server_cfg[&#34;auto_load_on_os&#34;] = auto_load_on_os
        mcp_servers[name] = server_cfg
    ai_settings[&#34;mcp_servers&#34;] = mcp_servers
    self.config.config[&#34;ai&#34;] = ai_settings
    self.config._saveconfig(self.config.file)</code></pre>
 </details>
 <div class="desc"><p>Update MCP server settings in the configuration with smart merging.</p></div>
 </dd>
 <dt id="connpy.services.ai_service.AIService.configure_provider"><code class="name flex">
 <span>def <span class="ident">configure_provider</span></span>(<span>self, provider, model=None, api_key=None)</span>
 </code></dt>
@@ -223,6 +566,75 @@ el.replaceWith(d);
 </details>
 <div class="desc"><p>Load a session's raw data by ID.</p></div>
 </dd>
 <dt id="connpy.services.ai_service.AIService.process_copilot_input"><code class="name flex">
 <span>def <span class="ident">process_copilot_input</span></span>(<span>self, input_text: str, session_state: dict) ‑> dict</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def process_copilot_input(self, input_text: str, session_state: dict) -&gt; dict:
    &#34;&#34;&#34;Parses slash commands and manages session state. Returns directive dict.&#34;&#34;&#34;
    text = input_text.strip()
    if not text.startswith(&#39;/&#39;):
        return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: text, &#34;overrides&#34;: {}}
    parts = text.split(maxsplit=1)
    cmd = parts[0].lower()
    args = parts[1] if len(parts) &gt; 1 else &#34;&#34;
    # 1. State Commands (Persistent)
    if cmd == &#34;/os&#34;:
        if args:
            session_state[&#39;os&#39;] = args
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: f&#34;OS context changed to {args}&#34;}
    elif cmd == &#34;/prompt&#34;:
        if args:
            session_state[&#39;prompt&#39;] = args
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: f&#34;Prompt regex changed to {args}&#34;}
    elif cmd == &#34;/memorize&#34;:
        if args:
            session_state[&#39;memories&#39;].append(args)
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: f&#34;Memory added: {args}&#34;}
    elif cmd == &#34;/clear&#34;:
        session_state[&#39;memories&#39;] = []
        return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Memory cleared&#34;}
    # 2. Hybrid Commands
    elif cmd == &#34;/architect&#34;:
        if not args:
            session_state[&#39;persona&#39;] = &#39;architect&#39;
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Persona set to Architect&#34;}
        else:
            return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;persona&#34;: &#34;architect&#34;}}
    elif cmd == &#34;/engineer&#34;:
        if not args:
            session_state[&#39;persona&#39;] = &#39;engineer&#39;
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Persona set to Engineer&#34;}
        else:
            return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;persona&#34;: &#34;engineer&#34;}}
    elif cmd == &#34;/trust&#34;:
        if not args:
            session_state[&#39;trust_mode&#39;] = True
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Auto-execute (trust) enabled for session&#34;}
        else:
            return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;trust&#34;: True}}
    elif cmd == &#34;/untrust&#34;:
        if not args:
            session_state[&#39;trust_mode&#39;] = False
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Auto-execute (trust) disabled for session&#34;}
        else:
            return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;trust&#34;: False}}
    # Unknown command, execute normally
    return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: text, &#34;overrides&#34;: {}}</code></pre>
 </details>
 <div class="desc"><p>Parses slash commands and manages session state. Returns directive dict.</p></div>
 </dd>
 </dl>
 <h3>Inherited members</h3>
 <ul class="hlist">
@@ -250,13 +662,18 @@ el.replaceWith(d);
 <ul>
 <li>
 <h4><code><a title="connpy.services.ai_service.AIService" href="#connpy.services.ai_service.AIService">AIService</a></code></h4>
-<ul class="two-column">
+<ul class="">
 <li><code><a title="connpy.services.ai_service.AIService.aask_copilot" href="#connpy.services.ai_service.AIService.aask_copilot">aask_copilot</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.ask" href="#connpy.services.ai_service.AIService.ask">ask</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.ask_copilot" href="#connpy.services.ai_service.AIService.ask_copilot">ask_copilot</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.build_context_blocks" href="#connpy.services.ai_service.AIService.build_context_blocks">build_context_blocks</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.configure_mcp" href="#connpy.services.ai_service.AIService.configure_mcp">configure_mcp</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.configure_provider" href="#connpy.services.ai_service.AIService.configure_provider">configure_provider</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.confirm" href="#connpy.services.ai_service.AIService.confirm">confirm</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.delete_session" href="#connpy.services.ai_service.AIService.delete_session">delete_session</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.list_sessions" href="#connpy.services.ai_service.AIService.list_sessions">list_sessions</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.load_session_data" href="#connpy.services.ai_service.AIService.load_session_data">load_session_data</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.process_copilot_input" href="#connpy.services.ai_service.AIService.process_copilot_input">process_copilot_input</a></code></li>
 </ul>
 </li>
 </ul>
@@ -265,7 +682,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.base API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -152,7 +152,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.config_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -117,6 +117,10 @@ el.replaceWith(d);
        if not isinstance(user_styles, dict):
            raise InvalidConfigurationError(&#34;Theme file must be a YAML dictionary.&#34;)
        # Support both direct styles and nested under &#39;theme&#39; key
        if &#34;theme&#34; in user_styles and isinstance(user_styles[&#34;theme&#34;], dict):
            user_styles = user_styles[&#34;theme&#34;]
        # Filter for valid styles only (prevent junk in config)
        valid_styles = {k: v for k, v in user_styles.items() if k in STYLES}
@@ -174,6 +178,10 @@ el.replaceWith(d);
    if not isinstance(user_styles, dict):
        raise InvalidConfigurationError(&#34;Theme file must be a YAML dictionary.&#34;)
    # Support both direct styles and nested under &#39;theme&#39; key
    if &#34;theme&#34; in user_styles and isinstance(user_styles[&#34;theme&#34;], dict):
        user_styles = user_styles[&#34;theme&#34;]
    # Filter for valid styles only (prevent junk in config)
    valid_styles = {k: v for k, v in user_styles.items() if k in STYLES}
@@ -311,7 +319,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.context_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -370,7 +370,7 @@ def current_context(self) -&gt; str:
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.exceptions API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -268,7 +268,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.execution_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -64,7 +64,7 @@ el.replaceWith(d);
        commands: List[str], 
        variables: Optional[Dict[str, Any]] = None,
        parallel: int = 10,
-        timeout: int = 10,
+        timeout: int = 20,
        folder: Optional[str] = None,
        prompt: Optional[str] = None,
        on_node_complete: Optional[Callable] = None,
@@ -112,7 +112,7 @@ el.replaceWith(d);
        expected: List[str],
        variables: Optional[Dict[str, Any]] = None,
        parallel: int = 10,
-        timeout: int = 10,
+        timeout: int = 20,
        folder: Optional[str] = None,
        prompt: Optional[str] = None,
        on_node_complete: Optional[Callable] = None,
@@ -189,7 +189,7 @@ el.replaceWith(d);
            &#34;commands&#34;: playbook[&#34;commands&#34;],
            &#34;variables&#34;: playbook.get(&#34;variables&#34;),
            &#34;parallel&#34;: options.get(&#34;parallel&#34;, parallel),
-            &#34;timeout&#34;: playbook.get(&#34;timeout&#34;, options.get(&#34;timeout&#34;, 10)),
+            &#34;timeout&#34;: playbook.get(&#34;timeout&#34;, options.get(&#34;timeout&#34;, 20)),
            &#34;prompt&#34;: options.get(&#34;prompt&#34;),
            &#34;name&#34;: playbook.get(&#34;name&#34;, &#34;Task&#34;)
        }
@@ -244,7 +244,7 @@ el.replaceWith(d);
 <div class="desc"><p>Run a plain-text script containing one command per line.</p></div>
 </dd>
 <dt id="connpy.services.execution_service.ExecutionService.run_commands"><code class="name flex">
-<span>def <span class="ident">run_commands</span></span>(<span>self,<br>nodes_filter: str,<br>commands: List[str],<br>variables: Dict[str, Any] | None = None,<br>parallel: int = 10,<br>timeout: int = 10,<br>folder: str | None = None,<br>prompt: str | None = None,<br>on_node_complete: Callable | None = None,<br>logger: Callable | None = None,<br>name: str | None = None) ‑> Dict[str, str]</span>
+<span>def <span class="ident">run_commands</span></span>(<span>self,<br>nodes_filter: str,<br>commands: List[str],<br>variables: Dict[str, Any] | None = None,<br>parallel: int = 10,<br>timeout: int = 20,<br>folder: str | None = None,<br>prompt: str | None = None,<br>on_node_complete: Callable | None = None,<br>logger: Callable | None = None,<br>name: str | None = None) ‑> Dict[str, str]</span>
 </code></dt>
 <dd>
 <details class="source">
@@ -257,7 +257,7 @@ el.replaceWith(d);
    commands: List[str], 
    variables: Optional[Dict[str, Any]] = None,
    parallel: int = 10,
-    timeout: int = 10,
+    timeout: int = 20,
    folder: Optional[str] = None,
    prompt: Optional[str] = None,
    on_node_complete: Optional[Callable] = None,
@@ -339,7 +339,7 @@ el.replaceWith(d);
        &#34;commands&#34;: playbook[&#34;commands&#34;],
        &#34;variables&#34;: playbook.get(&#34;variables&#34;),
        &#34;parallel&#34;: options.get(&#34;parallel&#34;, parallel),
-        &#34;timeout&#34;: playbook.get(&#34;timeout&#34;, options.get(&#34;timeout&#34;, 10)),
+        &#34;timeout&#34;: playbook.get(&#34;timeout&#34;, options.get(&#34;timeout&#34;, 20)),
        &#34;prompt&#34;: options.get(&#34;prompt&#34;),
        &#34;name&#34;: playbook.get(&#34;name&#34;, &#34;Task&#34;)
    }
@@ -360,7 +360,7 @@ el.replaceWith(d);
 <div class="desc"><p>Run a structured Connpy YAML automation playbook (from path or content).</p></div>
 </dd>
 <dt id="connpy.services.execution_service.ExecutionService.test_commands"><code class="name flex">
-<span>def <span class="ident">test_commands</span></span>(<span>self,<br>nodes_filter: str,<br>commands: List[str],<br>expected: List[str],<br>variables: Dict[str, Any] | None = None,<br>parallel: int = 10,<br>timeout: int = 10,<br>folder: str | None = None,<br>prompt: str | None = None,<br>on_node_complete: Callable | None = None,<br>logger: Callable | None = None,<br>name: str | None = None) ‑> Dict[str, Dict[str, bool]]</span>
+<span>def <span class="ident">test_commands</span></span>(<span>self,<br>nodes_filter: str,<br>commands: List[str],<br>expected: List[str],<br>variables: Dict[str, Any] | None = None,<br>parallel: int = 10,<br>timeout: int = 20,<br>folder: str | None = None,<br>prompt: str | None = None,<br>on_node_complete: Callable | None = None,<br>logger: Callable | None = None,<br>name: str | None = None) ‑> Dict[str, Dict[str, bool]]</span>
 </code></dt>
 <dd>
 <details class="source">
@@ -374,7 +374,7 @@ el.replaceWith(d);
    expected: List[str],
    variables: Optional[Dict[str, Any]] = None,
    parallel: int = 10,
-    timeout: int = 10,
+    timeout: int = 20,
    folder: Optional[str] = None,
    prompt: Optional[str] = None,
    on_node_complete: Optional[Callable] = None,
@@ -449,7 +449,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.import_export_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -361,7 +361,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -113,6 +113,138 @@ el.replaceWith(d);
 <pre><code class="python">class AIService(BaseService):
    &#34;&#34;&#34;Business logic for interacting with AI agents and LLM configurations.&#34;&#34;&#34;
    def build_context_blocks(self, raw_bytes: bytes, cmd_byte_positions: list, node_info: dict, last_line: str = &#34;&#34;) -&gt; list:
        &#34;&#34;&#34;Identifies command blocks in the terminal history.&#34;&#34;&#34;
        blocks = []
        if not raw_bytes:
            return blocks
        default_prompt = r&#39;&gt;$|#$|\$$|&gt;.$|#.$|\$.$&#39;
        device_prompt = node_info.get(&#34;prompt&#34;, default_prompt) if isinstance(node_info, dict) else default_prompt
        prompt_re_str = re.sub(r&#39;(?&lt;!\\)\$&#39;, &#39;&#39;, device_prompt)
        try:
            prompt_re = re.compile(prompt_re_str)
        except Exception:
            prompt_re = re.compile(re.sub(r&#39;(?&lt;!\\)\$&#39;, &#39;&#39;, default_prompt))
        parsed_positions = []
        if cmd_byte_positions and len(cmd_byte_positions) &gt;= 1:
            for i in range(1, len(cmd_byte_positions)):
                pos, known_cmd = cmd_byte_positions[i]
                prev_pos = cmd_byte_positions[i-1][0]
                if known_cmd:
                    prev_chunk = raw_bytes[prev_pos:pos]
                    prev_cleaned = log_cleaner(prev_chunk.decode(errors=&#39;replace&#39;))
                    prev_lines = [l for l in prev_cleaned.split(&#39;\n&#39;) if l.strip()]
                    prompt_text = prev_lines[-1].strip() if prev_lines else &#34;&#34;
                    preview = f&#34;{prompt_text}{known_cmd}&#34; if prompt_text else known_cmd
                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
                else:
                    chunk = raw_bytes[prev_pos:pos]
                    cleaned = log_cleaner(chunk.decode(errors=&#39;replace&#39;))
                    lines = [l for l in cleaned.split(&#39;\n&#39;) if l.strip()]
                    preview = lines[-1].strip() if lines else &#34;&#34;
                    if preview:
                        match = prompt_re.search(preview)
                        if match:
                            cmd_text = preview[match.end():].strip()
                            if cmd_text:
                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
                            else:
                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;EMPTY_PROMPT&#34;, &#34;preview&#34;: &#34;&#34;})
                        else:
                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
                    else:
                        parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
        last_newline = raw_bytes.rfind(b&#39;\n&#39;)
        current_prompt_pos = last_newline + 1 if last_newline != -1 else 0
        current_end = len(raw_bytes)
        for i, item in enumerate(parsed_positions):
            if item[&#34;type&#34;] == &#34;VALID_CMD&#34;:
                start_pos = item[&#34;pos&#34;]
                preview = item[&#34;preview&#34;]
                # Find the end position: next VALID_CMD or EMPTY_PROMPT
                end_pos = current_prompt_pos
                for j in range(i + 1, len(parsed_positions)):
                    next_item = parsed_positions[j]
                    if next_item[&#34;type&#34;] in (&#34;VALID_CMD&#34;, &#34;EMPTY_PROMPT&#34;):
                        end_pos = next_item[&#34;pos&#34;]
                        break
                blocks.append((start_pos, end_pos, preview))
        # Always ensure there is a final block representing the current prompt
        if not blocks:
            blocks.append((current_prompt_pos, current_end, last_line[:80] if last_line else &#34;CURRENT CONTEXT&#34;))
        elif blocks[-1][0] &lt; current_prompt_pos:
            blocks.append((current_prompt_pos, current_end, last_line[:80] if last_line else &#34;CURRENT CONTEXT&#34;))
        return blocks
    def process_copilot_input(self, input_text: str, session_state: dict) -&gt; dict:
        &#34;&#34;&#34;Parses slash commands and manages session state. Returns directive dict.&#34;&#34;&#34;
        text = input_text.strip()
        if not text.startswith(&#39;/&#39;):
            return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: text, &#34;overrides&#34;: {}}
        parts = text.split(maxsplit=1)
        cmd = parts[0].lower()
        args = parts[1] if len(parts) &gt; 1 else &#34;&#34;
        # 1. State Commands (Persistent)
        if cmd == &#34;/os&#34;:
            if args:
                session_state[&#39;os&#39;] = args
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: f&#34;OS context changed to {args}&#34;}
        elif cmd == &#34;/prompt&#34;:
            if args:
                session_state[&#39;prompt&#39;] = args
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: f&#34;Prompt regex changed to {args}&#34;}
        elif cmd == &#34;/memorize&#34;:
            if args:
                session_state[&#39;memories&#39;].append(args)
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: f&#34;Memory added: {args}&#34;}
        elif cmd == &#34;/clear&#34;:
            session_state[&#39;memories&#39;] = []
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Memory cleared&#34;}
        # 2. Hybrid Commands
        elif cmd == &#34;/architect&#34;:
            if not args:
                session_state[&#39;persona&#39;] = &#39;architect&#39;
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Persona set to Architect&#34;}
            else:
                return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;persona&#34;: &#34;architect&#34;}}
        elif cmd == &#34;/engineer&#34;:
            if not args:
                session_state[&#39;persona&#39;] = &#39;engineer&#39;
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Persona set to Engineer&#34;}
            else:
                return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;persona&#34;: &#34;engineer&#34;}}
        elif cmd == &#34;/trust&#34;:
            if not args:
                session_state[&#39;trust_mode&#39;] = True
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Auto-execute (trust) enabled for session&#34;}
            else:
                return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;trust&#34;: True}}
        elif cmd == &#34;/untrust&#34;:
            if not args:
                session_state[&#39;trust_mode&#39;] = False
                return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Auto-execute (trust) disabled for session&#34;}
            else:
                return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;trust&#34;: False}}
        # Unknown command, execute normally
        return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: text, &#34;overrides&#34;: {}}
    def ask(self, input_text, dryrun=False, chat_history=None, status=None, debug=False, session_id=None, console=None, chunk_callback=None, confirm_handler=None, trust=False, **overrides):
        &#34;&#34;&#34;Send a prompt to the AI agent.&#34;&#34;&#34;
        from connpy.ai import ai
@@ -126,6 +258,21 @@ el.replaceWith(d);
        agent = ai(self.config, console=console)
        return agent.confirm(input_text)
    def ask_copilot(self, terminal_buffer, user_question, node_info=None, chunk_callback=None):
        &#34;&#34;&#34;Ask the AI copilot for terminal assistance.&#34;&#34;&#34;
        from connpy.ai import ai, run_ai_async
        agent = ai(self.config)
        future = run_ai_async(agent.aask_copilot(terminal_buffer, user_question, node_info, chunk_callback=chunk_callback))
        return future.result()
    async def aask_copilot(self, terminal_buffer, user_question, node_info=None, chunk_callback=None):
        &#34;&#34;&#34;Ask the AI copilot for terminal assistance asynchronously.&#34;&#34;&#34;
        from connpy.ai import ai, run_ai_async
        import asyncio
        agent = ai(self.config)
        future = run_ai_async(agent.aask_copilot(terminal_buffer, user_question, node_info, chunk_callback=chunk_callback))
        return await asyncio.wrap_future(future)
    def list_sessions(self):
        &#34;&#34;&#34;Return a list of all saved AI sessions.&#34;&#34;&#34;
@@ -154,6 +301,40 @@ el.replaceWith(d);
        self.config.config[&#34;ai&#34;] = settings
        self.config._saveconfig(self.config.file)
    def configure_mcp(self, name, url=None, enabled=None, auto_load_on_os=None, remove=False):
        &#34;&#34;&#34;Update MCP server settings in the configuration with smart merging.&#34;&#34;&#34;
        ai_settings = self.config.config.get(&#34;ai&#34;, {})
        mcp_servers = ai_settings.get(&#34;mcp_servers&#34;, {})
        if remove:
            if name in mcp_servers:
                del mcp_servers[name]
        else:
            # Get existing or new
            server_cfg = mcp_servers.get(name, {})
            # Partial updates
            if url is not None:
                server_cfg[&#34;url&#34;] = url
            if enabled is not None:
                server_cfg[&#34;enabled&#34;] = bool(enabled)
            elif &#34;enabled&#34; not in server_cfg:
                server_cfg[&#34;enabled&#34;] = True # Default for new entries
            if auto_load_on_os is not None:
                if auto_load_on_os == &#34;&#34;: # Explicit clear
                    if &#34;auto_load_on_os&#34; in server_cfg:
                        del server_cfg[&#34;auto_load_on_os&#34;]
                else:
                    server_cfg[&#34;auto_load_on_os&#34;] = auto_load_on_os
            mcp_servers[name] = server_cfg
        ai_settings[&#34;mcp_servers&#34;] = mcp_servers
        self.config.config[&#34;ai&#34;] = ai_settings
        self.config._saveconfig(self.config.file)
    def load_session_data(self, session_id):
        &#34;&#34;&#34;Load a session&#39;s raw data by ID.&#34;&#34;&#34;
        from connpy.ai import ai
@@ -173,6 +354,24 @@ el.replaceWith(d);
 </ul>
 <h3>Methods</h3>
 <dl>
 <dt id="connpy.services.AIService.aask_copilot"><code class="name flex">
 <span>async def <span class="ident">aask_copilot</span></span>(<span>self, terminal_buffer, user_question, node_info=None, chunk_callback=None)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">async def aask_copilot(self, terminal_buffer, user_question, node_info=None, chunk_callback=None):
    &#34;&#34;&#34;Ask the AI copilot for terminal assistance asynchronously.&#34;&#34;&#34;
    from connpy.ai import ai, run_ai_async
    import asyncio
    agent = ai(self.config)
    future = run_ai_async(agent.aask_copilot(terminal_buffer, user_question, node_info, chunk_callback=chunk_callback))
    return await asyncio.wrap_future(future)</code></pre>
 </details>
 <div class="desc"><p>Ask the AI copilot for terminal assistance asynchronously.</p></div>
 </dd>
 <dt id="connpy.services.AIService.ask"><code class="name flex">
 <span>def <span class="ident">ask</span></span>(<span>self,<br>input_text,<br>dryrun=False,<br>chat_history=None,<br>status=None,<br>debug=False,<br>session_id=None,<br>console=None,<br>chunk_callback=None,<br>confirm_handler=None,<br>trust=False,<br>**overrides)</span>
 </code></dt>
@@ -189,6 +388,150 @@ el.replaceWith(d);
 </details>
 <div class="desc"><p>Send a prompt to the AI agent.</p></div>
 </dd>
 <dt id="connpy.services.AIService.ask_copilot"><code class="name flex">
 <span>def <span class="ident">ask_copilot</span></span>(<span>self, terminal_buffer, user_question, node_info=None, chunk_callback=None)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def ask_copilot(self, terminal_buffer, user_question, node_info=None, chunk_callback=None):
    &#34;&#34;&#34;Ask the AI copilot for terminal assistance.&#34;&#34;&#34;
    from connpy.ai import ai, run_ai_async
    agent = ai(self.config)
    future = run_ai_async(agent.aask_copilot(terminal_buffer, user_question, node_info, chunk_callback=chunk_callback))
    return future.result()</code></pre>
 </details>
 <div class="desc"><p>Ask the AI copilot for terminal assistance.</p></div>
 </dd>
 <dt id="connpy.services.AIService.build_context_blocks"><code class="name flex">
 <span>def <span class="ident">build_context_blocks</span></span>(<span>self,<br>raw_bytes: bytes,<br>cmd_byte_positions: list,<br>node_info: dict,<br>last_line: str = '') ‑> list</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def build_context_blocks(self, raw_bytes: bytes, cmd_byte_positions: list, node_info: dict, last_line: str = &#34;&#34;) -&gt; list:
    &#34;&#34;&#34;Identifies command blocks in the terminal history.&#34;&#34;&#34;
    blocks = []
    if not raw_bytes:
        return blocks
    default_prompt = r&#39;&gt;$|#$|\$$|&gt;.$|#.$|\$.$&#39;
    device_prompt = node_info.get(&#34;prompt&#34;, default_prompt) if isinstance(node_info, dict) else default_prompt
    prompt_re_str = re.sub(r&#39;(?&lt;!\\)\$&#39;, &#39;&#39;, device_prompt)
    try:
        prompt_re = re.compile(prompt_re_str)
    except Exception:
        prompt_re = re.compile(re.sub(r&#39;(?&lt;!\\)\$&#39;, &#39;&#39;, default_prompt))
    parsed_positions = []
    if cmd_byte_positions and len(cmd_byte_positions) &gt;= 1:
        for i in range(1, len(cmd_byte_positions)):
            pos, known_cmd = cmd_byte_positions[i]
            prev_pos = cmd_byte_positions[i-1][0]
            if known_cmd:
                prev_chunk = raw_bytes[prev_pos:pos]
                prev_cleaned = log_cleaner(prev_chunk.decode(errors=&#39;replace&#39;))
                prev_lines = [l for l in prev_cleaned.split(&#39;\n&#39;) if l.strip()]
                prompt_text = prev_lines[-1].strip() if prev_lines else &#34;&#34;
                preview = f&#34;{prompt_text}{known_cmd}&#34; if prompt_text else known_cmd
                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
            else:
                chunk = raw_bytes[prev_pos:pos]
                cleaned = log_cleaner(chunk.decode(errors=&#39;replace&#39;))
                lines = [l for l in cleaned.split(&#39;\n&#39;) if l.strip()]
                preview = lines[-1].strip() if lines else &#34;&#34;
                if preview:
                    match = prompt_re.search(preview)
                    if match:
                        cmd_text = preview[match.end():].strip()
                        if cmd_text:
                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
                        else:
                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;EMPTY_PROMPT&#34;, &#34;preview&#34;: &#34;&#34;})
                    else:
                        parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
                else:
                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
    last_newline = raw_bytes.rfind(b&#39;\n&#39;)
    current_prompt_pos = last_newline + 1 if last_newline != -1 else 0
    current_end = len(raw_bytes)
    for i, item in enumerate(parsed_positions):
        if item[&#34;type&#34;] == &#34;VALID_CMD&#34;:
            start_pos = item[&#34;pos&#34;]
            preview = item[&#34;preview&#34;]
            # Find the end position: next VALID_CMD or EMPTY_PROMPT
            end_pos = current_prompt_pos
            for j in range(i + 1, len(parsed_positions)):
                next_item = parsed_positions[j]
                if next_item[&#34;type&#34;] in (&#34;VALID_CMD&#34;, &#34;EMPTY_PROMPT&#34;):
                    end_pos = next_item[&#34;pos&#34;]
                    break
            blocks.append((start_pos, end_pos, preview))
    # Always ensure there is a final block representing the current prompt
    if not blocks:
        blocks.append((current_prompt_pos, current_end, last_line[:80] if last_line else &#34;CURRENT CONTEXT&#34;))
    elif blocks[-1][0] &lt; current_prompt_pos:
        blocks.append((current_prompt_pos, current_end, last_line[:80] if last_line else &#34;CURRENT CONTEXT&#34;))
    return blocks</code></pre>
 </details>
 <div class="desc"><p>Identifies command blocks in the terminal history.</p></div>
 </dd>
 <dt id="connpy.services.AIService.configure_mcp"><code class="name flex">
 <span>def <span class="ident">configure_mcp</span></span>(<span>self, name, url=None, enabled=None, auto_load_on_os=None, remove=False)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def configure_mcp(self, name, url=None, enabled=None, auto_load_on_os=None, remove=False):
    &#34;&#34;&#34;Update MCP server settings in the configuration with smart merging.&#34;&#34;&#34;
    ai_settings = self.config.config.get(&#34;ai&#34;, {})
    mcp_servers = ai_settings.get(&#34;mcp_servers&#34;, {})
    if remove:
        if name in mcp_servers:
            del mcp_servers[name]
    else:
        # Get existing or new
        server_cfg = mcp_servers.get(name, {})
        # Partial updates
        if url is not None:
            server_cfg[&#34;url&#34;] = url
        if enabled is not None:
            server_cfg[&#34;enabled&#34;] = bool(enabled)
        elif &#34;enabled&#34; not in server_cfg:
            server_cfg[&#34;enabled&#34;] = True # Default for new entries
        if auto_load_on_os is not None:
            if auto_load_on_os == &#34;&#34;: # Explicit clear
                if &#34;auto_load_on_os&#34; in server_cfg:
                    del server_cfg[&#34;auto_load_on_os&#34;]
            else:
                server_cfg[&#34;auto_load_on_os&#34;] = auto_load_on_os
        mcp_servers[name] = server_cfg
    ai_settings[&#34;mcp_servers&#34;] = mcp_servers
    self.config.config[&#34;ai&#34;] = ai_settings
    self.config._saveconfig(self.config.file)</code></pre>
 </details>
 <div class="desc"><p>Update MCP server settings in the configuration with smart merging.</p></div>
 </dd>
 <dt id="connpy.services.AIService.configure_provider"><code class="name flex">
 <span>def <span class="ident">configure_provider</span></span>(<span>self, provider, model=None, api_key=None)</span>
 </code></dt>
@@ -278,6 +621,75 @@ el.replaceWith(d);
 </details>
 <div class="desc"><p>Load a session's raw data by ID.</p></div>
 </dd>
 <dt id="connpy.services.AIService.process_copilot_input"><code class="name flex">
 <span>def <span class="ident">process_copilot_input</span></span>(<span>self, input_text: str, session_state: dict) ‑> dict</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def process_copilot_input(self, input_text: str, session_state: dict) -&gt; dict:
    &#34;&#34;&#34;Parses slash commands and manages session state. Returns directive dict.&#34;&#34;&#34;
    text = input_text.strip()
    if not text.startswith(&#39;/&#39;):
        return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: text, &#34;overrides&#34;: {}}
    parts = text.split(maxsplit=1)
    cmd = parts[0].lower()
    args = parts[1] if len(parts) &gt; 1 else &#34;&#34;
    # 1. State Commands (Persistent)
    if cmd == &#34;/os&#34;:
        if args:
            session_state[&#39;os&#39;] = args
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: f&#34;OS context changed to {args}&#34;}
    elif cmd == &#34;/prompt&#34;:
        if args:
            session_state[&#39;prompt&#39;] = args
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: f&#34;Prompt regex changed to {args}&#34;}
    elif cmd == &#34;/memorize&#34;:
        if args:
            session_state[&#39;memories&#39;].append(args)
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: f&#34;Memory added: {args}&#34;}
    elif cmd == &#34;/clear&#34;:
        session_state[&#39;memories&#39;] = []
        return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Memory cleared&#34;}
    # 2. Hybrid Commands
    elif cmd == &#34;/architect&#34;:
        if not args:
            session_state[&#39;persona&#39;] = &#39;architect&#39;
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Persona set to Architect&#34;}
        else:
            return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;persona&#34;: &#34;architect&#34;}}
    elif cmd == &#34;/engineer&#34;:
        if not args:
            session_state[&#39;persona&#39;] = &#39;engineer&#39;
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Persona set to Engineer&#34;}
        else:
            return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;persona&#34;: &#34;engineer&#34;}}
    elif cmd == &#34;/trust&#34;:
        if not args:
            session_state[&#39;trust_mode&#39;] = True
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Auto-execute (trust) enabled for session&#34;}
        else:
            return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;trust&#34;: True}}
    elif cmd == &#34;/untrust&#34;:
        if not args:
            session_state[&#39;trust_mode&#39;] = False
            return {&#34;action&#34;: &#34;state_update&#34;, &#34;message&#34;: &#34;Auto-execute (trust) disabled for session&#34;}
        else:
            return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: args, &#34;overrides&#34;: {&#34;trust&#34;: False}}
    # Unknown command, execute normally
    return {&#34;action&#34;: &#34;execute&#34;, &#34;clean_prompt&#34;: text, &#34;overrides&#34;: {}}</code></pre>
 </details>
 <div class="desc"><p>Parses slash commands and manages session state. Returns directive dict.</p></div>
 </dd>
 </dl>
 <h3>Inherited members</h3>
 <ul class="hlist">
@@ -359,6 +771,10 @@ el.replaceWith(d);
        if not isinstance(user_styles, dict):
            raise InvalidConfigurationError(&#34;Theme file must be a YAML dictionary.&#34;)
        # Support both direct styles and nested under &#39;theme&#39; key
        if &#34;theme&#34; in user_styles and isinstance(user_styles[&#34;theme&#34;], dict):
            user_styles = user_styles[&#34;theme&#34;]
        # Filter for valid styles only (prevent junk in config)
        valid_styles = {k: v for k, v in user_styles.items() if k in STYLES}
@@ -416,6 +832,10 @@ el.replaceWith(d);
    if not isinstance(user_styles, dict):
        raise InvalidConfigurationError(&#34;Theme file must be a YAML dictionary.&#34;)
    # Support both direct styles and nested under &#39;theme&#39; key
    if &#34;theme&#34; in user_styles and isinstance(user_styles[&#34;theme&#34;], dict):
        user_styles = user_styles[&#34;theme&#34;]
    # Filter for valid styles only (prevent junk in config)
    valid_styles = {k: v for k, v in user_styles.items() if k in STYLES}
@@ -590,7 +1010,7 @@ el.replaceWith(d);
        commands: List[str], 
        variables: Optional[Dict[str, Any]] = None,
        parallel: int = 10,
-        timeout: int = 10,
+        timeout: int = 20,
        folder: Optional[str] = None,
        prompt: Optional[str] = None,
        on_node_complete: Optional[Callable] = None,
@@ -638,7 +1058,7 @@ el.replaceWith(d);
        expected: List[str],
        variables: Optional[Dict[str, Any]] = None,
        parallel: int = 10,
-        timeout: int = 10,
+        timeout: int = 20,
        folder: Optional[str] = None,
        prompt: Optional[str] = None,
        on_node_complete: Optional[Callable] = None,
@@ -715,7 +1135,7 @@ el.replaceWith(d);
            &#34;commands&#34;: playbook[&#34;commands&#34;],
            &#34;variables&#34;: playbook.get(&#34;variables&#34;),
            &#34;parallel&#34;: options.get(&#34;parallel&#34;, parallel),
-            &#34;timeout&#34;: playbook.get(&#34;timeout&#34;, options.get(&#34;timeout&#34;, 10)),
+            &#34;timeout&#34;: playbook.get(&#34;timeout&#34;, options.get(&#34;timeout&#34;, 20)),
            &#34;prompt&#34;: options.get(&#34;prompt&#34;),
            &#34;name&#34;: playbook.get(&#34;name&#34;, &#34;Task&#34;)
        }
@@ -770,7 +1190,7 @@ el.replaceWith(d);
 <div class="desc"><p>Run a plain-text script containing one command per line.</p></div>
 </dd>
 <dt id="connpy.services.ExecutionService.run_commands"><code class="name flex">
-<span>def <span class="ident">run_commands</span></span>(<span>self,<br>nodes_filter: str,<br>commands: List[str],<br>variables: Dict[str, Any] | None = None,<br>parallel: int = 10,<br>timeout: int = 10,<br>folder: str | None = None,<br>prompt: str | None = None,<br>on_node_complete: Callable | None = None,<br>logger: Callable | None = None,<br>name: str | None = None) ‑> Dict[str, str]</span>
+<span>def <span class="ident">run_commands</span></span>(<span>self,<br>nodes_filter: str,<br>commands: List[str],<br>variables: Dict[str, Any] | None = None,<br>parallel: int = 10,<br>timeout: int = 20,<br>folder: str | None = None,<br>prompt: str | None = None,<br>on_node_complete: Callable | None = None,<br>logger: Callable | None = None,<br>name: str | None = None) ‑> Dict[str, str]</span>
 </code></dt>
 <dd>
 <details class="source">
@@ -783,7 +1203,7 @@ el.replaceWith(d);
    commands: List[str], 
    variables: Optional[Dict[str, Any]] = None,
    parallel: int = 10,
-    timeout: int = 10,
+    timeout: int = 20,
    folder: Optional[str] = None,
    prompt: Optional[str] = None,
    on_node_complete: Optional[Callable] = None,
@@ -865,7 +1285,7 @@ el.replaceWith(d);
        &#34;commands&#34;: playbook[&#34;commands&#34;],
        &#34;variables&#34;: playbook.get(&#34;variables&#34;),
        &#34;parallel&#34;: options.get(&#34;parallel&#34;, parallel),
-        &#34;timeout&#34;: playbook.get(&#34;timeout&#34;, options.get(&#34;timeout&#34;, 10)),
+        &#34;timeout&#34;: playbook.get(&#34;timeout&#34;, options.get(&#34;timeout&#34;, 20)),
        &#34;prompt&#34;: options.get(&#34;prompt&#34;),
        &#34;name&#34;: playbook.get(&#34;name&#34;, &#34;Task&#34;)
    }
@@ -886,7 +1306,7 @@ el.replaceWith(d);
 <div class="desc"><p>Run a structured Connpy YAML automation playbook (from path or content).</p></div>
 </dd>
 <dt id="connpy.services.ExecutionService.test_commands"><code class="name flex">
-<span>def <span class="ident">test_commands</span></span>(<span>self,<br>nodes_filter: str,<br>commands: List[str],<br>expected: List[str],<br>variables: Dict[str, Any] | None = None,<br>parallel: int = 10,<br>timeout: int = 10,<br>folder: str | None = None,<br>prompt: str | None = None,<br>on_node_complete: Callable | None = None,<br>logger: Callable | None = None,<br>name: str | None = None) ‑> Dict[str, Dict[str, bool]]</span>
+<span>def <span class="ident">test_commands</span></span>(<span>self,<br>nodes_filter: str,<br>commands: List[str],<br>expected: List[str],<br>variables: Dict[str, Any] | None = None,<br>parallel: int = 10,<br>timeout: int = 20,<br>folder: str | None = None,<br>prompt: str | None = None,<br>on_node_complete: Callable | None = None,<br>logger: Callable | None = None,<br>name: str | None = None) ‑> Dict[str, Dict[str, bool]]</span>
 </code></dt>
 <dd>
 <details class="source">
@@ -900,7 +1320,7 @@ el.replaceWith(d);
    expected: List[str],
    variables: Optional[Dict[str, Any]] = None,
    parallel: int = 10,
-    timeout: int = 10,
+    timeout: int = 20,
    folder: Optional[str] = None,
    prompt: Optional[str] = None,
    on_node_complete: Optional[Callable] = None,
@@ -2231,14 +2651,26 @@ el.replaceWith(d);
        from rich.console import Console
        from rich.console import Console
-        buf = io.StringIO()
+        import queue
        import threading
        q = queue.Queue()
        class QueueIO(io.StringIO):
            def write(self, s):
                q.put(s)
                return len(s)
            def flush(self):
                pass
        buf = QueueIO()
        old_console = printer._get_console()
        old_err_console = printer._get_err_console()
        def run_plugin():
            printer.set_thread_console(Console(file=buf, theme=printer.connpy_theme, force_terminal=True))
            printer.set_thread_err_console(Console(file=buf, theme=printer.connpy_theme, force_terminal=True))
            printer.set_thread_stream(buf)
            try:
                if hasattr(module, &#34;Entrypoint&#34;):
                    module.Entrypoint(args, parser, app)
@@ -2250,9 +2682,16 @@ el.replaceWith(d);
                printer.set_thread_console(old_console)
                printer.set_thread_err_console(old_err_console)
                printer.set_thread_stream(None)
                q.put(None)
-        for line in buf.getvalue().splitlines(keepends=True):
+        t = threading.Thread(target=run_plugin, daemon=True)
-            yield line</code></pre>
+        t.start()
        while True:
            item = q.get()
            if item is None:
                break
            yield item</code></pre>
 </details>
 <div class="desc"><p>Business logic for enabling, disabling, and listing plugins.</p>
 <p>Initialize the service.</p>
@@ -2507,14 +2946,26 @@ el.replaceWith(d);
    from rich.console import Console
    from rich.console import Console
-    buf = io.StringIO()
+    import queue
    import threading
    q = queue.Queue()
    class QueueIO(io.StringIO):
        def write(self, s):
            q.put(s)
            return len(s)
        def flush(self):
            pass
    buf = QueueIO()
    old_console = printer._get_console()
    old_err_console = printer._get_err_console()
    def run_plugin():
        printer.set_thread_console(Console(file=buf, theme=printer.connpy_theme, force_terminal=True))
        printer.set_thread_err_console(Console(file=buf, theme=printer.connpy_theme, force_terminal=True))
        printer.set_thread_stream(buf)
        try:
            if hasattr(module, &#34;Entrypoint&#34;):
                module.Entrypoint(args, parser, app)
@@ -2526,9 +2977,16 @@ el.replaceWith(d);
            printer.set_thread_console(old_console)
            printer.set_thread_err_console(old_err_console)
            printer.set_thread_stream(None)
            q.put(None)
-    for line in buf.getvalue().splitlines(keepends=True):
+    t = threading.Thread(target=run_plugin, daemon=True)
-        yield line</code></pre>
+    t.start()
    while True:
        item = q.get()
        if item is None:
            break
        yield item</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
@@ -3259,13 +3717,18 @@ el.replaceWith(d);
 <ul>
 <li>
 <h4><code><a title="connpy.services.AIService" href="#connpy.services.AIService">AIService</a></code></h4>
-<ul class="two-column">
+<ul class="">
 <li><code><a title="connpy.services.AIService.aask_copilot" href="#connpy.services.AIService.aask_copilot">aask_copilot</a></code></li>
 <li><code><a title="connpy.services.AIService.ask" href="#connpy.services.AIService.ask">ask</a></code></li>
 <li><code><a title="connpy.services.AIService.ask_copilot" href="#connpy.services.AIService.ask_copilot">ask_copilot</a></code></li>
 <li><code><a title="connpy.services.AIService.build_context_blocks" href="#connpy.services.AIService.build_context_blocks">build_context_blocks</a></code></li>
 <li><code><a title="connpy.services.AIService.configure_mcp" href="#connpy.services.AIService.configure_mcp">configure_mcp</a></code></li>
 <li><code><a title="connpy.services.AIService.configure_provider" href="#connpy.services.AIService.configure_provider">configure_provider</a></code></li>
 <li><code><a title="connpy.services.AIService.confirm" href="#connpy.services.AIService.confirm">confirm</a></code></li>
 <li><code><a title="connpy.services.AIService.delete_session" href="#connpy.services.AIService.delete_session">delete_session</a></code></li>
 <li><code><a title="connpy.services.AIService.list_sessions" href="#connpy.services.AIService.list_sessions">list_sessions</a></code></li>
 <li><code><a title="connpy.services.AIService.load_session_data" href="#connpy.services.AIService.load_session_data">load_session_data</a></code></li>
 <li><code><a title="connpy.services.AIService.process_copilot_input" href="#connpy.services.AIService.process_copilot_input">process_copilot_input</a></code></li>
 </ul>
 </li>
 <li>
@@ -3377,7 +3840,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.node_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -786,7 +786,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.plugin_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -284,14 +284,26 @@ el.replaceWith(d);
        from rich.console import Console
        from rich.console import Console
-        buf = io.StringIO()
+        import queue
        import threading
        q = queue.Queue()
        class QueueIO(io.StringIO):
            def write(self, s):
                q.put(s)
                return len(s)
            def flush(self):
                pass
        buf = QueueIO()
        old_console = printer._get_console()
        old_err_console = printer._get_err_console()
        def run_plugin():
            printer.set_thread_console(Console(file=buf, theme=printer.connpy_theme, force_terminal=True))
            printer.set_thread_err_console(Console(file=buf, theme=printer.connpy_theme, force_terminal=True))
            printer.set_thread_stream(buf)
            try:
                if hasattr(module, &#34;Entrypoint&#34;):
                    module.Entrypoint(args, parser, app)
@@ -303,9 +315,16 @@ el.replaceWith(d);
                printer.set_thread_console(old_console)
                printer.set_thread_err_console(old_err_console)
                printer.set_thread_stream(None)
                q.put(None)
-        for line in buf.getvalue().splitlines(keepends=True):
+        t = threading.Thread(target=run_plugin, daemon=True)
-            yield line</code></pre>
+        t.start()
        while True:
            item = q.get()
            if item is None:
                break
            yield item</code></pre>
 </details>
 <div class="desc"><p>Business logic for enabling, disabling, and listing plugins.</p>
 <p>Initialize the service.</p>
@@ -560,14 +579,26 @@ el.replaceWith(d);
    from rich.console import Console
    from rich.console import Console
-    buf = io.StringIO()
+    import queue
    import threading
    q = queue.Queue()
    class QueueIO(io.StringIO):
        def write(self, s):
            q.put(s)
            return len(s)
        def flush(self):
            pass
    buf = QueueIO()
    old_console = printer._get_console()
    old_err_console = printer._get_err_console()
    def run_plugin():
        printer.set_thread_console(Console(file=buf, theme=printer.connpy_theme, force_terminal=True))
        printer.set_thread_err_console(Console(file=buf, theme=printer.connpy_theme, force_terminal=True))
        printer.set_thread_stream(buf)
        try:
            if hasattr(module, &#34;Entrypoint&#34;):
                module.Entrypoint(args, parser, app)
@@ -579,9 +610,16 @@ el.replaceWith(d);
            printer.set_thread_console(old_console)
            printer.set_thread_err_console(old_err_console)
            printer.set_thread_stream(None)
            q.put(None)
-    for line in buf.getvalue().splitlines(keepends=True):
+    t = threading.Thread(target=run_plugin, daemon=True)
-        yield line</code></pre>
+    t.start()
    while True:
        item = q.get()
        if item is None:
            break
        yield item</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
@@ -671,7 +709,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.profile_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -429,7 +429,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.provider API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -164,7 +164,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.sync_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -964,7 +964,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.services.system_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -325,7 +325,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.conftest API documentation</title>
 <meta name="description" content="Shared fixtures for connpy tests …">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -258,7 +258,7 @@ def tmp_config_dir(tmp_path):
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -48,6 +48,10 @@ el.replaceWith(d);
 <dd>
 <div class="desc"><p>Tests for connpy.ai module.</p></div>
 </dd>
 <dt><code class="name"><a title="connpy.tests.test_ai_copilot" href="test_ai_copilot.html">connpy.tests.test_ai_copilot</a></code></dt>
 <dd>
 <div class="desc"></div>
 </dd>
 <dt><code class="name"><a title="connpy.tests.test_capture" href="test_capture.html">connpy.tests.test_capture</a></code></dt>
 <dd>
 <div class="desc"><p>Tests for connpy.core_plugins.capture</p></div>
@@ -131,6 +135,7 @@ el.replaceWith(d);
 <ul>
 <li><code><a title="connpy.tests.conftest" href="conftest.html">connpy.tests.conftest</a></code></li>
 <li><code><a title="connpy.tests.test_ai" href="test_ai.html">connpy.tests.test_ai</a></code></li>
 <li><code><a title="connpy.tests.test_ai_copilot" href="test_ai_copilot.html">connpy.tests.test_ai_copilot</a></code></li>
 <li><code><a title="connpy.tests.test_capture" href="test_capture.html">connpy.tests.test_capture</a></code></li>
 <li><code><a title="connpy.tests.test_completion" href="test_completion.html">connpy.tests.test_completion</a></code></li>
 <li><code><a title="connpy.tests.test_configfile" href="test_configfile.html">connpy.tests.test_configfile</a></code></li>
@@ -152,7 +157,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_ai API documentation</title>
 <meta name="description" content="Tests for connpy.ai module.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -1731,7 +1731,7 @@ def myai(self, ai_config, mock_litellm):
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -0,0 +1,315 @@
 <!doctype html>
 <html lang="en">
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
 <meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_ai_copilot API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/typography.min.css" integrity="sha512-Y1DYSb995BAfxobCkKepB1BqJJTPrOp3zPL74AWFugHHmmdcvO+C48WLrUOlhGMc0QG7AE3f7gmvvcrmX2fDoA==" crossorigin>
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/styles/default.min.css" crossorigin>
 <style>:root{--highlight-color:#fe9}.flex{display:flex !important}body{line-height:1.5em}#content{padding:20px}#sidebar{padding:1.5em;overflow:hidden}#sidebar > *:last-child{margin-bottom:2cm}.http-server-breadcrumbs{font-size:130%;margin:0 0 15px 0}#footer{font-size:.75em;padding:5px 30px;border-top:1px solid #ddd;text-align:right}#footer p{margin:0 0 0 1em;display:inline-block}#footer p:last-child{margin-right:30px}h1,h2,h3,h4,h5{font-weight:300}h1{font-size:2.5em;line-height:1.1em}h2{font-size:1.75em;margin:2em 0 .50em 0}h3{font-size:1.4em;margin:1.6em 0 .7em 0}h4{margin:0;font-size:105%}h1:target,h2:target,h3:target,h4:target,h5:target,h6:target{background:var(--highlight-color);padding:.2em 0}a{color:#058;text-decoration:none;transition:color .2s ease-in-out}a:visited{color:#503}a:hover{color:#b62}.title code{font-weight:bold}h2[id^="header-"]{margin-top:2em}.ident{color:#900;font-weight:bold}pre code{font-size:.8em;line-height:1.4em;padding:1em;display:block}code{background:#f3f3f3;font-family:"DejaVu Sans Mono",monospace;padding:1px 4px;overflow-wrap:break-word}h1 code{background:transparent}pre{border-top:1px solid #ccc;border-bottom:1px solid #ccc;margin:1em 0}#http-server-module-list{display:flex;flex-flow:column}#http-server-module-list div{display:flex}#http-server-module-list dt{min-width:10%}#http-server-module-list p{margin-top:0}.toc ul,#index{list-style-type:none;margin:0;padding:0}#index code{background:transparent}#index h3{border-bottom:1px solid #ddd}#index ul{padding:0}#index h4{margin-top:.6em;font-weight:bold}@media (min-width:200ex){#index .two-column{column-count:2}}@media (min-width:300ex){#index .two-column{column-count:3}}dl{margin-bottom:2em}dl dl:last-child{margin-bottom:4em}dd{margin:0 0 1em 3em}#header-classes + dl > dd{margin-bottom:3em}dd dd{margin-left:2em}dd p{margin:10px 0}.name{background:#eee;font-size:.85em;padding:5px 10px;display:inline-block;min-width:40%}.name:hover{background:#e0e0e0}dt:target .name{background:var(--highlight-color)}.name > span:first-child{white-space:nowrap}.name.class > span:nth-child(2){margin-left:.4em}.inherited{color:#999;border-left:5px solid #eee;padding-left:1em}.inheritance em{font-style:normal;font-weight:bold}.desc h2{font-weight:400;font-size:1.25em}.desc h3{font-size:1em}.desc dt code{background:inherit}.source > summary,.git-link-div{color:#666;text-align:right;font-weight:400;font-size:.8em;text-transform:uppercase}.source summary > *{white-space:nowrap;cursor:pointer}.git-link{color:inherit;margin-left:1em}.source pre{max-height:500px;overflow:auto;margin:0}.source pre code{font-size:12px;overflow:visible;min-width:max-content}.hlist{list-style:none}.hlist li{display:inline}.hlist li:after{content:',\2002'}.hlist li:last-child:after{content:none}.hlist .hlist{display:inline;padding-left:1em}img{max-width:100%}td{padding:0 .5em}.admonition{padding:.1em 1em;margin:1em 0}.admonition-title{font-weight:bold}.admonition.note,.admonition.info,.admonition.important{background:#aef}.admonition.todo,.admonition.versionadded,.admonition.tip,.admonition.hint{background:#dfd}.admonition.warning,.admonition.versionchanged,.admonition.deprecated{background:#fd4}.admonition.error,.admonition.danger,.admonition.caution{background:lightpink}</style>
 <style media="screen and (min-width: 700px)">@media screen and (min-width:700px){#sidebar{width:30%;height:100vh;overflow:auto;position:sticky;top:0}#content{width:70%;max-width:100ch;padding:3em 4em;border-left:1px solid #ddd}pre code{font-size:1em}.name{font-size:1em}main{display:flex;flex-direction:row-reverse;justify-content:flex-end}.toc ul ul,#index ul ul{padding-left:1em}.toc > ul > li{margin-top:.5em}}</style>
 <style media="print">@media print{#sidebar h1{page-break-before:always}.source{display:none}}@media print{*{background:transparent !important;color:#000 !important;box-shadow:none !important;text-shadow:none !important}a[href]:after{content:" (" attr(href) ")";font-size:90%}a[href][title]:after{content:none}abbr[title]:after{content:" (" attr(title) ")"}.ir a:after,a[href^="javascript:"]:after,a[href^="#"]:after{content:""}pre,blockquote{border:1px solid #999;page-break-inside:avoid}thead{display:table-header-group}tr,img{page-break-inside:avoid}img{max-width:100% !important}@page{margin:0.5cm}p,h2,h3{orphans:3;widows:3}h1,h2,h3,h4,h5,h6{page-break-after:avoid}}</style>
 <script defer src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/highlight.min.js" integrity="sha512-D9gUyxqja7hBtkWpPWGt9wfbfaMGVt9gnyCvYa+jojwwPHLCzUm5i8rpk7vD7wNee9bA35eYIjobYPaQuKS1MQ==" crossorigin></script>
 <script>window.addEventListener('DOMContentLoaded', () => {
 hljs.configure({languages: ['bash', 'css', 'diff', 'graphql', 'ini', 'javascript', 'json', 'plaintext', 'python', 'python-repl', 'rust', 'shell', 'sql', 'typescript', 'xml', 'yaml']});
 hljs.highlightAll();
 /* Collapse source docstrings */
 setTimeout(() => {
 [...document.querySelectorAll('.hljs.language-python > .hljs-string')]
 .filter(el => el.innerHTML.length > 200 && ['"""', "'''"].includes(el.innerHTML.substring(0, 3)))
 .forEach(el => {
 let d = document.createElement('details');
 d.classList.add('hljs-string');
 d.innerHTML = '<summary>"""</summary>' + el.innerHTML.substring(3);
 el.replaceWith(d);
 });
 }, 100);
 })</script>
 </head>
 <body>
 <main>
 <article id="content">
 <header>
 <h1 class="title">Module <code>connpy.tests.test_ai_copilot</code></h1>
 </header>
 <section id="section-intro">
 </section>
 <section>
 </section>
 <section>
 </section>
 <section>
 <h2 class="section-title" id="header-functions">Functions</h2>
 <dl>
 <dt id="connpy.tests.test_ai_copilot.mock_acompletion"><code class="name flex">
 <span>def <span class="ident">mock_acompletion</span></span>(<span>)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">@pytest.fixture
 def mock_acompletion():
    # Patch acompletion inside connpy.ai.aask_copilot
    with patch(&#39;litellm.acompletion&#39;) as mock:
        yield mock</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
 <dt id="connpy.tests.test_ai_copilot.test_aask_copilot_fallback"><code class="name flex">
 <span>def <span class="ident">test_aask_copilot_fallback</span></span>(<span>mock_acompletion)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def test_aask_copilot_fallback(mock_acompletion):
    agent = ai(DummyConfig())
    # Setup mock response for streaming
    class MockDelta:
        def __init__(self, content):
            self.content = content
    class MockChoice:
        def __init__(self, content):
            self.delta = MockDelta(content)
    class MockChunk:
        def __init__(self, content):
            self.choices = [MockChoice(content)]
    async def mock_ac(*args, **kwargs):
        return MockAsyncIterator([
            MockChunk(&#34;Here is some text response instead of tool call.&#34;)
        ])
    mock_acompletion.side_effect = mock_ac
    async def run_test():
        return await agent.aask_copilot(&#34;Router#&#34;, &#34;What do I do?&#34;)
    result = asyncio.run(run_test())
    if result[&#34;error&#34;]:
        print(f&#34;ERROR OCCURRED: {result[&#39;error&#39;]}&#34;)
    assert result[&#34;error&#34;] is None
    assert result[&#34;guide&#34;] == &#34;Here is some text response instead of tool call.&#34;
    assert result[&#34;risk_level&#34;] == &#34;low&#34;</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
 <dt id="connpy.tests.test_ai_copilot.test_aask_copilot_tool_call"><code class="name flex">
 <span>def <span class="ident">test_aask_copilot_tool_call</span></span>(<span>mock_acompletion)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def test_aask_copilot_tool_call(mock_acompletion):
    agent = ai(DummyConfig())
    # Setup mock response for streaming
    class MockDelta:
        def __init__(self, content):
            self.content = content
    class MockChoice:
        def __init__(self, content):
            self.delta = MockDelta(content)
    class MockChunk:
        def __init__(self, content):
            self.choices = [MockChoice(content)]
    # acompletion is awaited and returns an async iterator
    async def mock_ac(*args, **kwargs):
        return MockAsyncIterator([
            MockChunk(&#34;&lt;guide&gt;Check the interfaces and running config.&lt;/guide&gt;&#34;),
            MockChunk(&#34;&lt;commands&gt;\nshow ip int br\nshow run\n&lt;/commands&gt;&#34;),
            MockChunk(&#34;&lt;risk&gt;low&lt;/risk&gt;&#34;)
        ])
    mock_acompletion.side_effect = mock_ac
    async def run_test():
        return await agent.aask_copilot(&#34;Router#&#34;, &#34;What do I do?&#34;)
    result = asyncio.run(run_test())
    if result[&#34;error&#34;]:
        print(f&#34;ERROR OCCURRED: {result[&#39;error&#39;]}&#34;)
    assert result[&#34;error&#34;] is None
    assert result[&#34;guide&#34;] == &#34;Check the interfaces and running config.&#34;
    assert result[&#34;risk_level&#34;] == &#34;low&#34;
    assert result[&#34;commands&#34;] == [&#34;show ip int br&#34;, &#34;show run&#34;]</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
 <dt id="connpy.tests.test_ai_copilot.test_ingress_task_interception"><code class="name flex">
 <span>def <span class="ident">test_ingress_task_interception</span></span>(<span>)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def test_ingress_task_interception():
    async def run_test():
        c = node(&#34;test_node&#34;, &#34;1.2.3.4&#34;)
        c.mylog = MagicMock()
        c.mylog.getvalue.return_value = b&#34;Some session log&#34;
        c.unique = &#34;test_node&#34;
        c.host = &#34;1.2.3.4&#34;
        c.tags = {&#34;os&#34;: &#34;cisco_ios&#34;}
        class MockStream:
            def __init__(self):
                self.data = [b&#34;a&#34;, b&#34;b&#34;, b&#34;\x00&#34;, b&#34;c&#34;, b&#34;&#34;]
            async def read(self):
                if self.data:
                    return self.data.pop(0)
                return b&#34;&#34;
            def setup(self, resize_callback):
                pass
        stream = MockStream()
        called_copilot = False
        async def mock_handler(buffer, node_info, s, child_fd):
            nonlocal called_copilot
            called_copilot = True
            assert buffer == &#34;Some session log&#34;
            assert node_info[&#34;os&#34;] == &#34;cisco_ios&#34;
        c.child = MagicMock()
        c.child.child_fd = 123
        c.child.after = b&#34;&#34;
        c.child.buffer = b&#34;&#34;
        async def mock_ingress():
            while True:
                data = await stream.read()
                if not data:
                    break
                if mock_handler and b&#39;\x00&#39; in data:
                    buffer = c.mylog.getvalue().decode()
                    node_info = {&#34;name&#34;: getattr(c, &#39;unique&#39;, &#39;unknown&#39;), &#34;host&#34;: getattr(c, &#39;host&#39;, &#39;unknown&#39;)}
                    if isinstance(getattr(c, &#39;tags&#39;, None), dict):
                        node_info[&#34;os&#34;] = c.tags.get(&#34;os&#34;, &#34;unknown&#34;)
                    await mock_handler(buffer, node_info, stream, c.child.child_fd)
                    continue
        await mock_ingress()
        assert called_copilot
    asyncio.run(run_test())</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
 <dt id="connpy.tests.test_ai_copilot.test_logclean_ansi"><code class="name flex">
 <span>def <span class="ident">test_logclean_ansi</span></span>(<span>)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def test_logclean_ansi():
    c = node(&#34;test_node&#34;, &#34;1.2.3.4&#34;)
    raw = &#34;Router#\x1b[K\x1b[m show ip&#34;
    clean = c._logclean(raw, var=True)
    assert &#34;\x1b&#34; not in clean</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
 </dl>
 </section>
 <section>
 <h2 class="section-title" id="header-classes">Classes</h2>
 <dl>
 <dt id="connpy.tests.test_ai_copilot.DummyConfig"><code class="flex name class">
 <span>class <span class="ident">DummyConfig</span></span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">class DummyConfig:
    def __init__(self):
        self.config = {&#34;ai&#34;: {&#34;engineer_api_key&#34;: &#34;test_key&#34;, &#34;engineer_model&#34;: &#34;test_model&#34;}}
        self.defaultdir = &#34;/tmp&#34;</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
 <dt id="connpy.tests.test_ai_copilot.MockAsyncIterator"><code class="flex name class">
 <span>class <span class="ident">MockAsyncIterator</span></span>
 <span>(</span><span>items)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">class MockAsyncIterator:
    def __init__(self, items):
        self.items = items
    def __aiter__(self):
        return self
    async def __anext__(self):
        if not self.items:
            raise StopAsyncIteration
        return self.items.pop(0)</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
 </dl>
 </section>
 </article>
 <nav id="sidebar">
 <div class="toc">
 <ul></ul>
 </div>
 <ul id="index">
 <li><h3>Super-module</h3>
 <ul>
 <li><code><a title="connpy.tests" href="index.html">connpy.tests</a></code></li>
 </ul>
 </li>
 <li><h3><a href="#header-functions">Functions</a></h3>
 <ul class="">
 <li><code><a title="connpy.tests.test_ai_copilot.mock_acompletion" href="#connpy.tests.test_ai_copilot.mock_acompletion">mock_acompletion</a></code></li>
 <li><code><a title="connpy.tests.test_ai_copilot.test_aask_copilot_fallback" href="#connpy.tests.test_ai_copilot.test_aask_copilot_fallback">test_aask_copilot_fallback</a></code></li>
 <li><code><a title="connpy.tests.test_ai_copilot.test_aask_copilot_tool_call" href="#connpy.tests.test_ai_copilot.test_aask_copilot_tool_call">test_aask_copilot_tool_call</a></code></li>
 <li><code><a title="connpy.tests.test_ai_copilot.test_ingress_task_interception" href="#connpy.tests.test_ai_copilot.test_ingress_task_interception">test_ingress_task_interception</a></code></li>
 <li><code><a title="connpy.tests.test_ai_copilot.test_logclean_ansi" href="#connpy.tests.test_ai_copilot.test_logclean_ansi">test_logclean_ansi</a></code></li>
 </ul>
 </li>
 <li><h3><a href="#header-classes">Classes</a></h3>
 <ul>
 <li>
 <h4><code><a title="connpy.tests.test_ai_copilot.DummyConfig" href="#connpy.tests.test_ai_copilot.DummyConfig">DummyConfig</a></code></h4>
 </li>
 <li>
 <h4><code><a title="connpy.tests.test_ai_copilot.MockAsyncIterator" href="#connpy.tests.test_ai_copilot.MockAsyncIterator">MockAsyncIterator</a></code></h4>
 </li>
 </ul>
 </li>
 </ul>
 </nav>
 </main>
 <footer id="footer">
 <p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_capture API documentation</title>
 <meta name="description" content="Tests for connpy.core_plugins.capture">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -245,7 +245,7 @@ def mock_connapp():
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_completion API documentation</title>
 <meta name="description" content="Tests for connpy.completion module.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -257,7 +257,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_configfile API documentation</title>
 <meta name="description" content="Tests for connpy.configfile module.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -2005,7 +2005,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_connapp API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -699,7 +699,7 @@ def test_run(mock_run_commands, app):
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_core API documentation</title>
 <meta name="description" content="Tests for connpy.core module — node and nodes classes.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -1369,7 +1369,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_execution_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -142,7 +142,7 @@ Regression: ExecutionService.test_commands currently ignores on_node_complete.</
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_grpc_layer API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -581,8 +581,8 @@ def test_interact_node_uses_passed_name(self, mock_node, servicer):
        mock_resp = MagicMock()
        mock_resp.success = True
        mock_resp.stdout_data = b&#39;&#39;
        stub.stub.interact_node.return_value = iter([mock_resp])        
        with patch(&#34;connpy.printer.success&#34;) as mock_success:
            with patch(&#34;sys.stdin.fileno&#34;, return_value=0):
                mock_select.return_value = ([], [], [])
@@ -626,8 +626,8 @@ def test_connect_dynamic_msg_formatting_ssm(self, mock_select, mock_read, mock_s
    mock_resp = MagicMock()
    mock_resp.success = True
    mock_resp.stdout_data = b&#39;&#39;
    stub.stub.interact_node.return_value = iter([mock_resp])        
    with patch(&#34;connpy.printer.success&#34;) as mock_success:
        with patch(&#34;sys.stdin.fileno&#34;, return_value=0):
            mock_select.return_value = ([], [], [])
@@ -709,7 +709,7 @@ def test_connect_dynamic_msg_formatting_ssm(self, mock_select, mock_read, mock_s
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_hooks API documentation</title>
 <meta name="description" content="Tests for connpy.hooks module — MethodHook and ClassHook.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -673,7 +673,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_node_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -178,7 +178,7 @@ Regression: connapp._mod calls add_node instead of update_node.</p></div>
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_plugins API documentation</title>
 <meta name="description" content="Tests for connpy.plugins module.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -917,7 +917,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_printer API documentation</title>
 <meta name="description" content="Tests for connpy.printer module.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -459,7 +459,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_printer_concurrency API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -148,7 +148,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_profile_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -192,7 +192,7 @@ Regression: ProfileService currently doesn't resolve inheritance within profiles
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_provider API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -139,7 +139,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tests.test_sync API documentation</title>
 <meta name="description" content="Tests for connpy.services.sync_service">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -354,7 +354,7 @@ def test_perform_restore(self, mock_remove, mock_dirname, mock_exists, MockZipFi
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.6">
+<meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.tunnels API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -94,6 +94,24 @@ el.replaceWith(d);
                # signal handling not supported on some loops (e.g., Windows Proactor)
                pass
    def stop_reading(self):
        &#34;&#34;&#34;Temporarily stop reading from stdin.&#34;&#34;&#34;
        if self._loop and self.stdin_fd is not None:
            try:
                self._loop.remove_reader(self.stdin_fd)
            except Exception:
                pass
    def start_reading(self):
        &#34;&#34;&#34;Resume reading from stdin.&#34;&#34;&#34;
        if self._loop and self.stdin_fd is not None:
            try:
                # Ensure we don&#39;t add it twice
                self._loop.remove_reader(self.stdin_fd)
            except Exception:
                pass
            self._loop.add_reader(self.stdin_fd, self._read_ready)
    def teardown(self):
        if self._loop:
            try:
@@ -216,6 +234,44 @@ Handles terminal raw mode, async I/O, and SIGWINCH signals.</p></div>
 </details>
 <div class="desc"></div>
 </dd>
 <dt id="connpy.tunnels.LocalStream.start_reading"><code class="name flex">
 <span>def <span class="ident">start_reading</span></span>(<span>self)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def start_reading(self):
    &#34;&#34;&#34;Resume reading from stdin.&#34;&#34;&#34;
    if self._loop and self.stdin_fd is not None:
        try:
            # Ensure we don&#39;t add it twice
            self._loop.remove_reader(self.stdin_fd)
        except Exception:
            pass
        self._loop.add_reader(self.stdin_fd, self._read_ready)</code></pre>
 </details>
 <div class="desc"><p>Resume reading from stdin.</p></div>
 </dd>
 <dt id="connpy.tunnels.LocalStream.stop_reading"><code class="name flex">
 <span>def <span class="ident">stop_reading</span></span>(<span>self)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def stop_reading(self):
    &#34;&#34;&#34;Temporarily stop reading from stdin.&#34;&#34;&#34;
    if self._loop and self.stdin_fd is not None:
        try:
            self._loop.remove_reader(self.stdin_fd)
        except Exception:
            pass</code></pre>
 </details>
 <div class="desc"><p>Temporarily stop reading from stdin.</p></div>
 </dd>
 <dt id="connpy.tunnels.LocalStream.teardown"><code class="name flex">
 <span>def <span class="ident">teardown</span></span>(<span>self)</span>
 </code></dt>
@@ -293,6 +349,7 @@ Handles terminal raw mode, async I/O, and SIGWINCH signals.</p></div>
        self.response_queue = response_queue
        self.running = True
        self._reader_queue = asyncio.Queue()
        self.copilot_queue = asyncio.Queue()
        self.resize_callback = None
        self._loop = None
        self.t = None
@@ -309,6 +366,19 @@ Handles terminal raw mode, async I/O, and SIGWINCH signals.</p></div>
                    if req.cols &gt; 0 and req.rows &gt; 0:
                        if self.resize_callback:
                            self._loop.call_soon_threadsafe(self.resize_callback, req.rows, req.cols)
                    # Copilot dispatching
                    copilot_msg = {}
                    if getattr(req, &#34;copilot_question&#34;, &#34;&#34;):
                        copilot_msg.update({
                            &#34;question&#34;: req.copilot_question,
                            &#34;context_buffer&#34;: getattr(req, &#34;copilot_context_buffer&#34;, &#34;&#34;),
                            &#34;node_info_json&#34;: getattr(req, &#34;copilot_node_info_json&#34;, &#34;&#34;)
                        })
                    if getattr(req, &#34;copilot_action&#34;, &#34;&#34;):
                        copilot_msg[&#34;action&#34;] = req.copilot_action
                    if copilot_msg:
                        self._loop.call_soon_threadsafe(self.copilot_queue.put_nowait, copilot_msg)
                    if req.stdin_data:
                        self._loop.call_soon_threadsafe(self._reader_queue.put_nowait, req.stdin_data)
            except Exception:
@@ -374,6 +444,19 @@ Bridges the blocking gRPC iterators with the async _async_interact_loop.</p></di
                if req.cols &gt; 0 and req.rows &gt; 0:
                    if self.resize_callback:
                        self._loop.call_soon_threadsafe(self.resize_callback, req.rows, req.cols)
                # Copilot dispatching
                copilot_msg = {}
                if getattr(req, &#34;copilot_question&#34;, &#34;&#34;):
                    copilot_msg.update({
                        &#34;question&#34;: req.copilot_question,
                        &#34;context_buffer&#34;: getattr(req, &#34;copilot_context_buffer&#34;, &#34;&#34;),
                        &#34;node_info_json&#34;: getattr(req, &#34;copilot_node_info_json&#34;, &#34;&#34;)
                    })
                if getattr(req, &#34;copilot_action&#34;, &#34;&#34;):
                    copilot_msg[&#34;action&#34;] = req.copilot_action
                if copilot_msg:
                    self._loop.call_soon_threadsafe(self.copilot_queue.put_nowait, copilot_msg)
                if req.stdin_data:
                    self._loop.call_soon_threadsafe(self._reader_queue.put_nowait, req.stdin_data)
        except Exception:
@@ -438,9 +521,11 @@ Bridges the blocking gRPC iterators with the async _async_interact_loop.</p></di
 <ul>
 <li>
 <h4><code><a title="connpy.tunnels.LocalStream" href="#connpy.tunnels.LocalStream">LocalStream</a></code></h4>
-<ul class="">
+<ul class="two-column">
 <li><code><a title="connpy.tunnels.LocalStream.read" href="#connpy.tunnels.LocalStream.read">read</a></code></li>
 <li><code><a title="connpy.tunnels.LocalStream.setup" href="#connpy.tunnels.LocalStream.setup">setup</a></code></li>
 <li><code><a title="connpy.tunnels.LocalStream.start_reading" href="#connpy.tunnels.LocalStream.start_reading">start_reading</a></code></li>
 <li><code><a title="connpy.tunnels.LocalStream.stop_reading" href="#connpy.tunnels.LocalStream.stop_reading">stop_reading</a></code></li>
 <li><code><a title="connpy.tunnels.LocalStream.teardown" href="#connpy.tunnels.LocalStream.teardown">teardown</a></code></li>
 <li><code><a title="connpy.tunnels.LocalStream.write" href="#connpy.tunnels.LocalStream.write">write</a></code></li>
 </ul>
@@ -460,7 +545,7 @@ Bridges the blocking gRPC iterators with the async _async_interact_loop.</p></di
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -0,0 +1,150 @@
 <!doctype html>
 <html lang="en">
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
 <meta name="generator" content="pdoc3 0.11.5">
 <title>connpy.utils API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/typography.min.css" integrity="sha512-Y1DYSb995BAfxobCkKepB1BqJJTPrOp3zPL74AWFugHHmmdcvO+C48WLrUOlhGMc0QG7AE3f7gmvvcrmX2fDoA==" crossorigin>
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/styles/default.min.css" crossorigin>
 <style>:root{--highlight-color:#fe9}.flex{display:flex !important}body{line-height:1.5em}#content{padding:20px}#sidebar{padding:1.5em;overflow:hidden}#sidebar > *:last-child{margin-bottom:2cm}.http-server-breadcrumbs{font-size:130%;margin:0 0 15px 0}#footer{font-size:.75em;padding:5px 30px;border-top:1px solid #ddd;text-align:right}#footer p{margin:0 0 0 1em;display:inline-block}#footer p:last-child{margin-right:30px}h1,h2,h3,h4,h5{font-weight:300}h1{font-size:2.5em;line-height:1.1em}h2{font-size:1.75em;margin:2em 0 .50em 0}h3{font-size:1.4em;margin:1.6em 0 .7em 0}h4{margin:0;font-size:105%}h1:target,h2:target,h3:target,h4:target,h5:target,h6:target{background:var(--highlight-color);padding:.2em 0}a{color:#058;text-decoration:none;transition:color .2s ease-in-out}a:visited{color:#503}a:hover{color:#b62}.title code{font-weight:bold}h2[id^="header-"]{margin-top:2em}.ident{color:#900;font-weight:bold}pre code{font-size:.8em;line-height:1.4em;padding:1em;display:block}code{background:#f3f3f3;font-family:"DejaVu Sans Mono",monospace;padding:1px 4px;overflow-wrap:break-word}h1 code{background:transparent}pre{border-top:1px solid #ccc;border-bottom:1px solid #ccc;margin:1em 0}#http-server-module-list{display:flex;flex-flow:column}#http-server-module-list div{display:flex}#http-server-module-list dt{min-width:10%}#http-server-module-list p{margin-top:0}.toc ul,#index{list-style-type:none;margin:0;padding:0}#index code{background:transparent}#index h3{border-bottom:1px solid #ddd}#index ul{padding:0}#index h4{margin-top:.6em;font-weight:bold}@media (min-width:200ex){#index .two-column{column-count:2}}@media (min-width:300ex){#index .two-column{column-count:3}}dl{margin-bottom:2em}dl dl:last-child{margin-bottom:4em}dd{margin:0 0 1em 3em}#header-classes + dl > dd{margin-bottom:3em}dd dd{margin-left:2em}dd p{margin:10px 0}.name{background:#eee;font-size:.85em;padding:5px 10px;display:inline-block;min-width:40%}.name:hover{background:#e0e0e0}dt:target .name{background:var(--highlight-color)}.name > span:first-child{white-space:nowrap}.name.class > span:nth-child(2){margin-left:.4em}.inherited{color:#999;border-left:5px solid #eee;padding-left:1em}.inheritance em{font-style:normal;font-weight:bold}.desc h2{font-weight:400;font-size:1.25em}.desc h3{font-size:1em}.desc dt code{background:inherit}.source > summary,.git-link-div{color:#666;text-align:right;font-weight:400;font-size:.8em;text-transform:uppercase}.source summary > *{white-space:nowrap;cursor:pointer}.git-link{color:inherit;margin-left:1em}.source pre{max-height:500px;overflow:auto;margin:0}.source pre code{font-size:12px;overflow:visible;min-width:max-content}.hlist{list-style:none}.hlist li{display:inline}.hlist li:after{content:',\2002'}.hlist li:last-child:after{content:none}.hlist .hlist{display:inline;padding-left:1em}img{max-width:100%}td{padding:0 .5em}.admonition{padding:.1em 1em;margin:1em 0}.admonition-title{font-weight:bold}.admonition.note,.admonition.info,.admonition.important{background:#aef}.admonition.todo,.admonition.versionadded,.admonition.tip,.admonition.hint{background:#dfd}.admonition.warning,.admonition.versionchanged,.admonition.deprecated{background:#fd4}.admonition.error,.admonition.danger,.admonition.caution{background:lightpink}</style>
 <style media="screen and (min-width: 700px)">@media screen and (min-width:700px){#sidebar{width:30%;height:100vh;overflow:auto;position:sticky;top:0}#content{width:70%;max-width:100ch;padding:3em 4em;border-left:1px solid #ddd}pre code{font-size:1em}.name{font-size:1em}main{display:flex;flex-direction:row-reverse;justify-content:flex-end}.toc ul ul,#index ul ul{padding-left:1em}.toc > ul > li{margin-top:.5em}}</style>
 <style media="print">@media print{#sidebar h1{page-break-before:always}.source{display:none}}@media print{*{background:transparent !important;color:#000 !important;box-shadow:none !important;text-shadow:none !important}a[href]:after{content:" (" attr(href) ")";font-size:90%}a[href][title]:after{content:none}abbr[title]:after{content:" (" attr(title) ")"}.ir a:after,a[href^="javascript:"]:after,a[href^="#"]:after{content:""}pre,blockquote{border:1px solid #999;page-break-inside:avoid}thead{display:table-header-group}tr,img{page-break-inside:avoid}img{max-width:100% !important}@page{margin:0.5cm}p,h2,h3{orphans:3;widows:3}h1,h2,h3,h4,h5,h6{page-break-after:avoid}}</style>
 <script defer src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/highlight.min.js" integrity="sha512-D9gUyxqja7hBtkWpPWGt9wfbfaMGVt9gnyCvYa+jojwwPHLCzUm5i8rpk7vD7wNee9bA35eYIjobYPaQuKS1MQ==" crossorigin></script>
 <script>window.addEventListener('DOMContentLoaded', () => {
 hljs.configure({languages: ['bash', 'css', 'diff', 'graphql', 'ini', 'javascript', 'json', 'plaintext', 'python', 'python-repl', 'rust', 'shell', 'sql', 'typescript', 'xml', 'yaml']});
 hljs.highlightAll();
 /* Collapse source docstrings */
 setTimeout(() => {
 [...document.querySelectorAll('.hljs.language-python > .hljs-string')]
 .filter(el => el.innerHTML.length > 200 && ['"""', "'''"].includes(el.innerHTML.substring(0, 3)))
 .forEach(el => {
 let d = document.createElement('details');
 d.classList.add('hljs-string');
 d.innerHTML = '<summary>"""</summary>' + el.innerHTML.substring(3);
 el.replaceWith(d);
 });
 }, 100);
 })</script>
 </head>
 <body>
 <main>
 <article id="content">
 <header>
 <h1 class="title">Module <code>connpy.utils</code></h1>
 </header>
 <section id="section-intro">
 </section>
 <section>
 </section>
 <section>
 </section>
 <section>
 <h2 class="section-title" id="header-functions">Functions</h2>
 <dl>
 <dt id="connpy.utils.log_cleaner"><code class="name flex">
 <span>def <span class="ident">log_cleaner</span></span>(<span>data: str) ‑> str</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">def log_cleaner(data: str) -&gt; str:
    &#34;&#34;&#34;
    Stateless utility to remove ANSI sequences and process cursor movements.
    &#34;&#34;&#34;
    if not data:
        return &#34;&#34;
    lines = data.split(&#39;\n&#39;)
    cleaned_lines = []
    # Regex to capture: ANSI sequences, control characters (\r, \b, etc), and plain text chunks
    token_re = re.compile(r&#39;(\x1B(?:[@-Z\\-_]|\[[0-?]*[ -/ ]*[@-~])|\r|\b|\x7f|[\x00-\x1F]|[^\x1B\r\b\x7f\x00-\x1F]+)&#39;)
    for line in lines:
        buffer = []
        cursor = 0
        for token in token_re.findall(line):
            if token == &#39;\r&#39;:
                cursor = 0
            elif token in (&#39;\b&#39;, &#39;\x7f&#39;):
                if cursor &gt; 0:
                    cursor -= 1
            elif token.startswith(&#39;\x1B[&#39;) and len(token) &gt;= 3:
                # Parse CSI: \x1B[ &lt;params&gt; &lt;final_char&gt;
                final = token[-1]
                param_str = token[2:-1]
                n = int(param_str) if param_str.isdigit() else 1
                if final == &#39;D&#39;:   # CUB – Cursor Back
                    cursor = max(0, cursor - n)
                elif final == &#39;C&#39;: # CUF – Cursor Forward
                    cursor = min(len(buffer), cursor + n)
                elif final == &#39;K&#39;: # EL  – Erase in Line
                    if n == 0 or param_str == &#39;&#39;:      # Clear to end
                        buffer = buffer[:cursor]
                    elif n == 1:                         # Clear to start
                        buffer[:cursor] = [&#39; &#39;] * cursor
                    elif n == 2:                         # Clear entire line
                        buffer = []
                        cursor = 0
                elif final == &#39;G&#39;: # CHA – Cursor Horizontal Absolute (1-indexed)
                    cursor = max(0, n - 1)
                    # Pad buffer if cursor is beyond current length
                    if cursor &gt; len(buffer):
                        buffer.extend([&#39; &#39;] * (cursor - len(buffer)))
                elif final == &#39;P&#39;: # DCH – Delete Characters
                    del buffer[cursor:cursor + n]
                elif final == &#39;@&#39;: # ICH – Insert Characters
                    buffer[cursor:cursor] = [&#39; &#39;] * n
                # All other CSI sequences are silently discarded
            elif token.startswith(&#39;\x1B&#39;):
                continue
            elif len(token) == 1 and ord(token) &lt; 32:
                continue
            else:
                for char in token:
                    if cursor == len(buffer):
                        buffer.append(char)
                    else:
                        buffer[cursor] = char
                    cursor += 1
        cleaned_lines.append(&#34;&#34;.join(buffer))
    return &#34;\n&#34;.join(cleaned_lines).replace(&#39;\n\n&#39;, &#39;\n&#39;).strip()</code></pre>
 </details>
 <div class="desc"><p>Stateless utility to remove ANSI sequences and process cursor movements.</p></div>
 </dd>
 </dl>
 </section>
 <section>
 </section>
 </article>
 <nav id="sidebar">
 <div class="toc">
 <ul></ul>
 </div>
 <ul id="index">
 <li><h3>Super-module</h3>
 <ul>
 <li><code><a title="connpy" href="index.html">connpy</a></code></li>
 </ul>
 </li>
 <li><h3><a href="#header-functions">Functions</a></h3>
 <ul class="">
 <li><code><a title="connpy.utils.log_cleaner" href="#connpy.utils.log_cleaner">log_cleaner</a></code></li>
 </ul>
 </li>
 </ul>
 </nav>
 </main>
 <footer id="footer">
 <p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
 </footer>
 </body>
 </html>
@@ -13,5 +13,10 @@ protobuf>=6.31.1,<7.0.0
 google-api-python-client>=2.125.0
 google-auth-oauthlib>=1.2.0
 google-auth-httplib2>=0.2.0
 prompt-toolkit>=3.0.0
 mcp>=1.2.0
 aiohttp>=3.9.0
 httpx>=0.27.0
 requests>=2.31.0
 pytest>=8.0.0
 pytest-mock>=3.12.0
@@ -18,12 +18,16 @@ classifiers =
    Topic :: System :: Networking
    Intended Audience :: Telecommunications Industry
    Programming Language :: Python :: 3
    Programming Language :: Python :: 3.10
    Programming Language :: Python :: 3.11
    Programming Language :: Python :: 3.12
    Natural Language :: English
    Operating System :: MacOS
    Operating System :: Unix
 [options]
 packages = find:
 python_requires = >=3.10
 install_requires =
    rich>=13.7.1
    rich-argparse>=1.4.0
@@ -40,6 +44,11 @@ install_requires =
    google-api-python-client>=2.125.0
    google-auth-oauthlib>=1.2.0
    google-auth-httplib2>=0.2.0
    prompt-toolkit>=3.0.0
    mcp>=1.2.0
    aiohttp>=3.9.0
    httpx>=0.27.0
    requests>=2.31.0
 [options.entry_points]
 console_scripts =
@@ -51,4 +60,4 @@ console_scripts =
 connpy = 
    core_plugins/*
    proto/*
-    grpc/*.proto
+    grpc_layer/*
Author	SHA1	Message	Date
fluzzi32	3ca6f497d1	bug fix missing manifest	2026-05-18 16:57:02 -03:00
fluzzi32	1db4a045e5	bug fix double sys import	2026-05-18 15:46:21 -03:00
fluzzi32	7b01a0c05f	update pdoc	2026-05-18 14:11:28 -03:00
fluzzi32	5a8b744aa8	📌 1. Uncommitted Changes (Staged for next commit) Focus: UI stability and efficient rendering. * Markdown Rendering Rewrite: Removed the dependency on rich.live.Live (which caused flickering and high CPU usage by constantly re-rendering the entire panel). * New BlockMarkdownRenderer: Implemented in printer.py (alias IncrementalMarkdownParser), it accumulates text in a buffer and only prints to the screen when it detects a complete block (e.g., line breaks or ` code blocks). * UI Optimizations (terminal_ui.py & stubs.py): Waiting spinners now stop cleanly, and the UI transitions smoothly to block printing. Fixed visual truncation issues in the bottom "Tab prompt" bar for excessively long commands. 📦 2. Commit History (Last 4 Commits) `9446baf` - improve ai rules * Strict Anti-Hallucination (ai.py): Injected MANDATORY rules into the System Prompt for both Architect and Engineer agents. Now, if the terminal buffer is empty or only contains an idle prompt (e.g., iol#), the AI is strictly forced to state that it lacks data instead of inventing topologies or configurations. * Language Preference: Explicitly instructed agents to always respond in the same language the user used to ask the question. `64377f7` - move context block logic to server and improvements * Context Precision (ai_service.py): Moved context partitioning logic to the service/server side. It now calculates exact start_pos and end_pos based on identified commands, preventing mixed outputs or residual text from bleeding into the AI's prompt. * Token Savings (server.py): The server now selectively strips garbage metadata and UI caches that add no value to the LLM before sending the payload over the wire. `e4fd1ad` - fix logclean for 6wind * Full ANSI/CSI Support (utils.py): Replaced the legacy rigid escape filter with a complete CSI (Control Sequence Introducer) parser. The client can now accurately process numeric cursor movements (C, D), inline dynamic erasures (K), and absolute shifts (G), ensuring connpy understands exactly what a 6WIND router or other VNFs render on the screen without garbage characters. `b0a914a` - updates al copilot * Copilot <> gRPC Integration: Implemented the async plumbing required for the Copilot to work over the gRPC tunnel (server.py & stubs.py). * Initial Streaming UI: Added the first iteration of chunk callbacks to provide real-time feedback while the LLM generates responses, laying the groundwork for the uncommitted block-renderer optimizations.	2026-05-18 14:07:24 -03:00
fluzzi32	9446bafc0c	improve ai rules	2026-05-15 23:28:22 -03:00
fluzzi32	64377f7f30	move context block logic to server and improvemnets	2026-05-15 18:24:48 -03:00
fluzzi32	e4fd1adba3	fix logclean for6wind	2026-05-15 17:32:09 -03:00
fluzzi32	b0a914ad7f	updates al copilot	2026-05-15 16:25:18 -03:00
fluzzi32	05fb9951c6	update readme	2026-05-13 14:49:04 -03:00
fluzzi32	12543c683e	1. Persistence Setup: Optimized the dockerfile to manually create the /root/.config/conn/.folder file pointing to /config. This avoids running the conn command during the build process and ensures a cleaner setup. 2. Copilot UI Fix: Resolved a double-escaping bug in the terminal bottom bar. Device prompts (like 6WIND-PE1>) will now render correctly instead of showing HTML entities like >. 3. AI Model Update: Updated the default engineer model in connpy/ai.py to gemini/gemini-3.1-flash-lite, removing the deprecated -preview suffix. 4. Standardized Timeouts: Unified all default timeouts to 20 seconds across the board. This includes direct execution (run/test), modern playbooks (v2), and classic task-based playbooks (v1). 5. Documentation Update: Regenerated the full documentation site in the docs/ directory using pdoc to reflect the latest changes. 6. Cleanup: Removed all debug prints from connpy/core.py and restored the docker/logs/.gitignore file.	2026-05-13 14:16:14 -03:00
fluzzi32	3ad4f6da1f	add utils file	2026-05-12 13:19:35 -03:00
fluzzi32	5c4e2116c8	refactor to support future command center	2026-05-12 13:19:06 -03:00
fluzzi32	2666a3db1f	feat(v6): integrate MCP support, async copilot streaming, and enhanced terminal UI logic	2026-05-12 12:20:50 -03:00
fluzzi32	dba7e24dda	feat(copilot): stabilize concurrency and enhance terminal context management - AI Concurrency: - Implemented a dedicated background event loop (ConnpyAILoop) in a separate thread for AI tasks to ensure thread safety and event loop affinity. - Added 'run_ai_async' utility to funnel all LiteLLM calls through the dedicated loop. - Implemented global 'cleanup()' for safe closure of sync/async LiteLLM sessions. - gRPC & Remote Sessions: - Enhanced 'NodeServicer' to identify command blocks within the terminal buffer using prompt regex/byte tracking. - Added support for selective context retrieval via 'context_start_pos' in the gRPC Interact stream. - Synchronized remote Copilot behavior by enriching questions with session history (last 5 queries) in 'NodeStub'. - Optimized token usage by cleaning 'node_info' metadata before AI transmission. - Terminal Context & Core: - Modified 'node.connect' to always initialize 'mylog' (BytesIO) buffer regardless of disk logging configuration, ensuring Copilot context availability. - Integrated 'ai.cleanup()' in CLI (connapp) and Server (api) exit points for graceful shutdowns. - Suppressed LiteLLM internal streaming coroutine warnings during task cancellation.	2026-05-11 12:30:43 -03:00
fluzzi32	1103393be6	minor bugs	2026-05-09 19:45:19 -03:00
fluzzi32	a08d84139d	remoto con fixes de todo	2026-05-08 21:53:57 -03:00
fluzzi32	d62af8b037	copilot remoto con bugs	2026-05-08 18:45:42 -03:00
fluzzi32	14a298dd4d	context ai	2026-05-07 17:30:43 -03:00
fluzzi32	61552cffa9	add stream	2026-05-07 16:43:32 -03:00
fluzzi32	c4d704f473	first implementation phase 1 and 2	2026-05-07 15:17:00 -03:00
fluzzi32	f6078b6fde	update version	2026-05-05 18:25:33 -03:00
fluzzi32	37db74f47d	refactor(core): stabilize gRPC streaming, plugin invocation, and CLI UX - Implement threaded plugin execution with Queue-based streaming in PluginService - Refactor remote logger to preserve ANSI colors and fix TTY line endings (\r\n) - Intelligent terminal filtering: disable SSM screen-clearing filter after success - Sanitize SSH-only flags in core.py when using SFTP protocol - Rewrite completion tree with pre/post-node states and flag deduplication - Update gRPC unit tests to match new streaming response structure	2026-05-05 18:24:31 -03:00
`@@ -1 +1 @@`
	`__version__ = "6.0.0b5"`	`__version__ = "6.0.0b11"`