mcpField
H
Hydra MCP

Hydra MCP

Local
github.com/keyboardsmokeDeveloper Tools3Updated 82w ago
Repository

About Hydra MCP

Hydra MCP is an intermediary server that coordinates functions registered by multiple clients and makes them available to Claude through a REST API and MCP interface. It includes sample client and Claude implementations for testing.

Key features

  1. 01Central registry for client functions
  2. 02REST API for discovery and invocation
  3. 03Register functions from clients
  4. 04Discover endpoints and functions
  5. 05Invoke remote client functions
  6. 06Sample implementations for testing

Use cases

  1. 01Expose functions from several clients to Claude
  2. 02Discover functions registered at an endpoint
  3. 03Route a function call to a client
  4. 04Test client registration with sample implementations

MCP support

Its MCP interface exposes tools to list endpoints, retrieve registered functions for an endpoint, and call a function with arguments. The central server accepts client function registrations and routes requests.

#ai#mcp#claude

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

Hydra MCP

This project implements a Hydra MCP (Model Control Protocol) server that enables Claude to interact with various client functions through a REST API interface. The system consists of a central server that coordinates communication between multiple clients and exposes their functions to Claude through a standardized API.

Claude using the Hydra MCP plugin

Project Structure

  • ipc_server.py: The central FastAPI server that manages function registration and routing
  • ipc_api.py: The Hydra MCP interface that exposes client functions to Claude
  • fake_client.py: A sample client implementation for testing
  • fake_claude.py: A test implementation of Claude's interface

Features

  • Centralized function registry for multiple clients
  • REST API endpoints for function discovery and invocation
  • Standardized communication protocol between clients and Claude
  • Easy-to-use function registration system
  • Automatic function discovery and documentation

API Endpoints

Server Endpoints
  • POST /register_functions: Register new functions from a client
  • GET /get_functions: Retrieve the list of all registered functions
MCP Tools
  • get_endpoints(): List all available endpoints
  • get_registered_functions(endpoint): Get functions available at a specific endpoint
  • call_function(endpoint, function_name, arguments): Call a specific function with arguments

Usage

  1. Start the central server:
python ipc_server.py
  1. Register client functions:
# Example client registration
functions = [
    {
        "name": "example_function",
        "description": "An example function",
        "parameters": {
            "param1": "string",
            "param2": "integer"
        }
    }
]

requests.post(
    f"http://localhost:{API_SERVER_PORT}/register_functions",
    json={
        "endpoint": "client1",
        "functions": functions
    }
)
  1. Use the MCP interface to interact with functions:
# Get available endpoints
endpoints = get_endpoints()

# Get functions for a specific endpoint
functions = get_registered_functions("localhost:1234")

# Call a function
result = call_function("localhost:1234", "example_function", {"param1": "value", "param2": 42})

Configuration

The server runs on port 6565 by default. This can be modified by changing the API_SERVER_PORT constant in both ipc_server.py and ipc_api.py.

Dependencies

  • FastAPI
  • uvicorn
  • requests
  • fastmcp

Testing

The project includes sample implementations (fake_client.py and fake_claude.py) for testing the functionality. These can be used as reference implementations for creating new clients or testing the system.

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.