drio
Open app

AgentRPC

Source

A universal RPC layer for AI agents. Connect to any function, any language, across network boundaries. AgentRPC wraps your functions in a universal RPC interface, connecting them to a hosted server accessible through open standards including MCP.

Catalog onlyCatalog onlySTDIO

Overview

AgentRPC is a universal RPC layer designed for AI agents, enabling connections to any function across different programming languages and network boundaries.

To use AgentRPC, obtain an API key, choose your preferred SDK (Go, Node.js, or Python), and register your functions using the provided SDKs. You can also set up an MCP server for external AI models to interact with your registered tools.

  • Multi-language support for TypeScript, Go, Python, and .NET (coming soon) - Private network support without the need for open ports - Long-running function support with long polling SDKs - Full observability with tracing, metrics, and events - Automatic failover and retries for enhanced reliability - Compatibility with MCP and OpenAI SDKs
  1. Connecting AI models deployed in private VPCs.
  2. Integrating various services across multiple cloud environments.
  3. Enabling long-running function calls beyond HTTP timeout limits.

Add to your AI client

Use these steps to connect AgentRPC in Cursor, Claude, VS Code, and other MCP-compatible apps. The same JSON appears in the Use with menu above for one-click copy.

Cursor

Add this to your .cursor/mcp.json file in your project root, then restart Cursor.

.cursor/mcp.json

{
  "mcpServers": {
    "agentrpc-agentrpc": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-agentrpc-agentrpc"
      ]
    }
  }
}

Claude Desktop

Add this server entry to the mcpServers object in your Claude Desktop config, then restart the app.

~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows)

{
  "mcpServers": {
    "agentrpc-agentrpc": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-agentrpc-agentrpc"
      ]
    }
  }
}

Claude Code

Add this to your project's .mcp.json file. Claude Code will detect it automatically.

.mcp.json (project root)

{
  "mcpServers": {
    "agentrpc-agentrpc": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-agentrpc-agentrpc"
      ]
    }
  }
}

VS Code (Copilot)

Add this to your .vscode/mcp.json file. Requires the GitHub Copilot extension with MCP support enabled.

.vscode/mcp.json

{
  "servers": {
    "agentrpc-agentrpc": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-agentrpc-agentrpc"
      ]
    }
  }
}

Windsurf

Add this to your Windsurf MCP config file, then restart Windsurf.

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "agentrpc-agentrpc": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-agentrpc-agentrpc"
      ]
    }
  }
}

Cline

Open Cline settings, navigate to MCP Servers, and add this server configuration.

Cline MCP Settings (via UI)

{
  "mcpServers": {
    "agentrpc-agentrpc": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-agentrpc-agentrpc"
      ]
    }
  }
}

FAQ

What programming languages does AgentRPC support?

AgentRPC supports TypeScript, Go, Python, and .NET (coming soon).

Is there a cost to use AgentRPC?

AgentRPC offers a free tier, but check the website for detailed pricing.

How do I monitor the health of my functions?

AgentRPC provides comprehensive tracing and metrics for full observability.7:["$","div",null,{"className":"container mx-auto flex flex-col gap-4","children":["$L26","$L27",["$","$L28",null,{"currentProject":{"id":3385,"uuid":"51899261-0d5c-48cb-84c1-dbaf744701a9","name":"agentrpc","title":"AgentRPC","description":"A universal RPC layer for AI agents. Connect to any function, any language, across network boundaries.\r\n\r\nAgentRPC wraps your functions in a universal RPC interface, connecting them to a hosted server accessible through open standards including MCP.","avatar_url":"https://avatars.githubusercontent.com/u/203365055","created_at":"2025-03-19T07:01:02.964Z","updated_at":"2025-03-19T07:02:57.039Z","status":"created","author_name":"agentrpc","author_avatar_url":null,"tags":"agentrpc,rpc-layer,ai-agents","category":"developer-tools","is_featured":false,"sort":1,"url":"https://github.com/agentrpc/agentrpc","target":"_self","content":"$29","summary":"$2a","img_url":null,"type":null,"metadata":null,"user_uuid":"07eb9aa8-6c5d-4549-ac16-5bf2cfa27247","tools":null,"sse_url":null,"sse_provider":null,"sse_params":null,"is_official":false,"server_command":null,"server_params":null,"server_config":null,"allow_call":false,"is_innovation":false,"is_dxt":false,"dxt_manifest":null,"dxt_file_url":null,"is_audit":false},"randomProjects":[],"currentServerKey":"$undefined"}]]}]

AgentRPC MCP Server — MCP Registry