ishevel/pydentycai-mcp-server

PydanticAI MCP Server in Go

This project is a lightweight MCP (Model Context Protocol) server written in Go, designed for remote management of PydanticAI agents. It allows starting, stopping, getting status, and updating the configuration of PydanticAI agents via a standardized MCP protocol.

Overview

The server provides a set of tools that enable interaction with PydanticAI agents, managing their lifecycle and configuration. Agent configurations are stored in text files (JSON) that are updated by MCP commands.

MCP Server Tools

The MCP server provides the following tools for managing PydanticAI agents:

• run_agent

  • Description: Starts a PydanticAI agent with a specified ID.
  • Input Parameters: agent_id (string, required) - Unique identifier for the agent to be launched.

• stop_agent

  • Description: Stops a running PydanticAI agent by its ID.
  • Input Parameters: agent_id (string, required) - Unique identifier for the agent to be stopped.

• update_agent_config

  • Description: Updates the configuration file for a PydanticAI agent.
  • Input Parameters: agent_id (string, required) - Unique identifier for the agent whose configuration needs to be updated; config_data (string, required) - Configuration data in JSON format (string).

• get_agent_status

  • Description: Returns the current status of a running PydanticAI agent by its ID.
  • Input Parameters: agent_id (string, required) - Unique identifier for the agent whose status is to be retrieved.

• list_agents

  • Description: Returns a list of all registered agents and their statuses.
  • Input Parameters: None.

Connecting to MCP Clients (e.g., Cline for VS Code)

To connect our MCP server to a client like Cline in VS Code, you need to add the corresponding configuration to your cline_mcp_settings.json file (typically located at C:/Users/vm-user/AppData/Roaming/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json).

Since our server uses the Stdio transport (standard input/output), the configuration will look as follows:

"pydentycai-mcp-server": {
  "autoApprove": [
    "run_agent",
    "stop_agent",
    "get_agent_status",
    "update_agent_config",
    "list_agents"
  ],
  "timeout": 60,
  "type": "stdio",
  "command": "./pydentycai-mcp-server.exe",
  "args": []
}

Configuration Explanation:

• "pydentycai-mcp-server": A unique name that will be used to identify our server in Cline.
• "autoApprove": A list of tools whose calls will be automatically approved by Cline. It is recommended to include all tools here to avoid constant confirmation prompts.
• "timeout": Maximum time in seconds to wait for a response from the server.
• "type": "stdio": Indicates that the server uses standard input/output for communication.
• "command": "./pydentycai-mcp-server.exe": The path to our Go server executable. It is assumed that the file is in the project's root directory.
• "args": []: Additional command-line arguments for launching the server (none in this case).

After adding this configuration to cline_mcp_settings.json and restarting VS Code (or performing the relevant action in Cline to refresh servers), you will be able to call our server's tools directly from Cline.

Running the Server

You can run the server using the following command in the project's root directory:

./pydentycai-mcp-server.exe

Development and Testing

For development and testing, you can use standard Go commands:

• go mod tidy: For managing dependencies.
• go build -o pydentycai-mcp-server.exe: For building the executable.