mcpField
M
MCP Agent Connector

MCP Agent Connector

Local
github.com/Sumedh1599AI & Agents0Updated 78w ago

About MCP Agent Connector

MCP Agent Connector is an open-source library for agent platforms that discovers generated MCP servers and keeps their capabilities synchronized from build metadata. It publishes server manifests and supports agent mapping and optional local-network discovery.

Key features

  1. 01Publish tool and resource manifests
  2. 02Support mcp-agent and Lutra
  3. 03Optional local-network metadata broadcast
  4. 04Map Claude agents from desktop config
  5. 05Sync capabilities with clients

Use cases

  1. 01Make generated MCP server capabilities discoverable
  2. 02Keep client environments updated with server changes
  3. 03Map Claude agents to MCP servers

MCP support

It publishes manifests of MCP server tools and resources for MCP-compliant agent platforms to discover. Optional local-network broadcasts and client synchronization provide additional discovery and update paths.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

MCP Agent Connector

GitHub GitHub last commit GitHub issues

Introduction

The MCP Agent Connector is a powerful open-source library designed to enable any MCP-compliant agentic platform (such as Claude, Grok, TypingMind, Cursor, etc.) to auto-discover generated MCP servers and sync capabilities directly from build metadata. This library streamlines the integration process by automating the discovery of MCP servers and ensuring real-time synchronization with client environments.

Features

  • Auto-publish MCP server manifest: Automatically publishes a manifest of tools and resources available on the MCP server.
  • Supports mcp-agent and Lutra agents: Compatible with both mcp-agent and Lutra agent platforms, offering flexible integration options.
  • Metadata broadcast over local network (optional): Optionally broadcasts metadata over the local network to facilitate automatic discovery by compatible agents.
  • Claude agent mapping via Claude Desktop config: Maps Claude agents using configurations from Claude Desktop for seamless integration.
  • Real-time sync with client environment: Ensures that client environments are kept up-to-date with the latest capabilities and information from MCP servers.

Installation Instructions

To install the mcp_agent_connector library, you can use pip:

pip install mcp-agent-connector

Alternatively, if you prefer to clone the repository and install it manually, follow these steps:

  1. Clone the repository:

    git clone https://github.com/your-username/mcp_agent_connector.git
    
  2. Navigate into the project directory:

    cd mcp_agent_connector
    
  3. Install the library using pip:

    pip install .
    

Usage Examples

Example 1: Basic Initialization and Auto-publishing
from mcp_agent_connector import MCPAgentConnector

# Initialize the connector with default settings
connector = MCPAgentConnector()

# Start auto-publishing the MCP server manifest
connector.start()
Example 2: Custom Configuration for Claude Agent Mapping
from mcp_agent_connector import MCPAgentConnector, ClaudeConfig

# Define custom configuration for Claude agents
claude_config = ClaudeConfig(
    desktop_path="/path/to/claude/desktop/config",
    agent_mapping={"agent1": "server1", "agent2": "server2"}
)

# Initialize the connector with the custom configuration
connector = MCPAgentConnector(claude_config=claude_config)

# Start auto-publishing and mapping Claude agents
connector.start()
Example 3: Enabling Metadata Broadcast Over Local Network
from mcp_agent_connector import MCPAgentConnector, BroadcastConfig

# Define broadcast configuration
broadcast_config = BroadcastConfig(
    enabled=True,
    network_interface="eth0",
    broadcast_interval=60  # seconds
)

# Initialize the connector with broadcast configuration
connector = MCPAgentConnector(broadcast_config=broadcast_config)

# Start broadcasting and auto-publishing
connector.start()

API Documentation

Class: MCPAgentConnector
  • __init__(self, claude_config=None, broadcast_config=None)

    • Initializes the MCP Agent Connector.
    • Parameters:
      • claude_config: (Optional) An instance of ClaudeConfig for Claude agent mapping.
      • broadcast_config: (Optional) An instance of BroadcastConfig to enable metadata broadcasting over the local network.
  • start(self)

    • Starts the auto-publishing process and any additional configured features such as Claude agent mapping or metadata broadcasting.
Class: ClaudeConfig
  • __init__(self, desktop_path=None, agent_mapping=None)
    • Initializes configuration for Claude agents.
    • Parameters:
      • desktop_path: (Optional) Path to the Claude Desktop configuration file.
      • agent_mapping: (Optional) A dictionary mapping Claude agent names to MCP server names.
Class: BroadcastConfig
  • __init__(self, enabled=False, network_interface=None, broadcast_interval=30)
    • Initializes configuration for local network metadata broadcasting.
    • Parameters:
      • enabled: (Default: False) Boolean indicating whether broadcasting is enabled.
      • network_interface: (Optional) The network interface to use for broadcasting.
      • broadcast_interval: (Default: 30 seconds) Interval between broadcasts.

License

This project is licensed under the MIT License. See the LICENSE file for details.


Feel free to contribute to this open-source library by submitting pull requests or reporting issues on GitHub. We welcome any feedback and suggestions to improve mcp_agent_connector!

⚠️ Development Status

This library is currently in early development. Some tests may be failing with the following issues:

Contributions to fix these issues are welcome! Please submit a pull request if you have a solution.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.