home / mcp / agentrpc mcp server

AgentRPC MCP Server

Exposes an MCP-compliant server to connect external AI models with registered tools.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "agentrpc-agentrpc": {
      "command": "npx",
      "args": [
        "-y",
        "agentrpc",
        "mcp"
      ],
      "env": {
        "AGENTRPC_API_SECRET": "<YOUR_API_SECRET>"
      }
    }
  }
}

You can run an MCP server with AgentRPC to expose your registered tools to external AI agents. This server acts as a bridge, enabling open, standardized interactions across private networks and multiple environments. It provides a hosted MCP interface and OpenAPI SDK compatible tool definitions so clients can connect securely and reliably.

How to use

Use the MCP server to expose your registered tools to external agents. You interact with it by running the MCP server locally or via integration configurations, then connect client tools that understand MCP for tool discovery and invocation. Your setup enables health monitoring, automatic retry, and observability across the network.

How to install

Prerequisites: you need Node.js and npm installed on your system to run the MCP server via npx. Ensure you have access to your API secret for AgentRPC.

ANGENTRPC_API_SECRET=YOUR_API_SECRET npx agentrpc mcp

Claude Desktop Integration

Configure Claude Desktop to connect to the MCP server by adding an MCP entry that runs the AgentRPC MCP server.

{
  "mcpServers": {
    "agentrpc": {
      "command": "npx",
      "args": [
        "-y",
        "agentrpc",
        "mcp"
      ],
      "env": {
        "AGENTRPC_API_SECRET": "<YOUR_API_SECRET>"
      }
    }
  }
}

Cursor Integration

Add an MCP server entry to your Cursor configuration to connect via the MCP interface.

{
  "mcpServers": {
    "agentrpc": {
      "command": "npx",
      "args": ["-y", "agentrpc", "mcp"],
      "env": {
        "AGENTRPC_API_SECRET": "<YOUR_API_SECRET>"
      }
    }
  }
}

Available tools

mcp

MCP server wrapper that enables external AI models to interact with registered tools via the Model Context Protocol.