AgentRPC

What is AgentRPC?

AgentRPC is a universal RPC (Remote Procedure Call) layer designed for AI agents. It allows agents to connect to and execute functions written in any programming language, across any network boundary, including private VPCs and Kubernetes clusters. This simplifies the integration of disparate services and tools into a cohesive AI-powered workflow.

How to use AgentRPC?

To use AgentRPC, you first register your functions or APIs using the AgentRPC SDK available for TypeScript, Go, and Python. The AgentRPC platform then manages these registered functions, monitors their health, and provides a hosted MCP server and OpenAPI SDK compatible tool definitions. This allows AI agents that support these standards (like those from OpenAI, Anthropic, or custom builds with LiteLLM) to easily discover and call the registered functions.

Key Features

• Multi-language Support: Connect to tools written in TypeScript, Go, Python, with .NET support coming soon.
• Private Network Support: Register and call functions in private VPCs without needing to open any ports.
• Long-running Functions: Supports long-polling to handle function calls that exceed standard HTTP timeout limits.
• Full Observability: Provides comprehensive tracing, metrics, and events for complete visibility into your RPC calls.
• Automatic Failover: Includes intelligent health tracking with automatic failover and retries to ensure reliability.
• Framework Compatibility: Offers out-of-the-box support for popular frameworks like Express.js, FastAPI, and Go's net/http.

Use Cases

• Integrating Microservices: An AI agent can orchestrate tasks across multiple microservices written in different languages and deployed in various environments.
• Securely Accessing Internal Tools: Allow an AI agent to securely interact with internal APIs and databases located in a private cloud without exposing them to the public internet.
• Building Complex AI Workflows: Create sophisticated workflows where an AI agent can delegate tasks to specialized functions, such as data processing, file manipulation, or interacting with third-party services, regardless of where those functions are hosted.

FAQ from the project

As a developer-focused tool, common questions would likely center on:

• Supported Languages and Frameworks: What languages are currently supported and what is the roadmap for new additions?
• Security Model: How does AgentRPC ensure secure communication between the agent and the functions, especially across network boundaries?
• Performance and Latency: What is the expected overhead or latency introduced by the AgentRPC layer?
• Pricing and Hosting: Is there a self-hosted option, and what does the pricing model for the managed service look like?

For detailed answers, developers should consult the official AgentRPC documentation and GitHub repository.

AgentRPC

Universal RPC layer for AI agents across network boundaries and languages

Overview

AgentRPC allows you to connect to any function, in any language, across network boundaries. It's ideal when you have services deployed in:

• Private VPCs
• Kubernetes clusters
• Multiple cloud environments

AgentRPC wraps your functions in a universal RPC interface, connecting them to a hosted RPC server accessible through open standards:

• Model Context Protocol (MCP)
• OpenAI-compatible tool definitions (OpenAI, Anthropic, LiteLLM, OpenRouter, etc.)

How It Works

1. Registration: Use our SDK to register functions and APIs in any language
2. Management: The AgentRPC platform (api.agentrpc.com) registers the function and monitors its health
3. Access: Receive OpenAPI SDK compatible tool definitions and a hosted MCP server for connecting to compatible agents

Key Features

Feature	Description
Multi-language Support	Connect to tools in TypeScript, Go, Python and .NET (coming soon)
Private Network Support	Register functions in private VPCs with no open ports required
Long-running Functions	Long polling SDKs allow function calls beyond HTTP timeout limits
Full Observability	Comprehensive tracing, metrics, and events for complete visibility
Automatic Failover	Intelligent health tracking with automatic failover and retries
Framework Compatibility	Out-of-the-box support for MCP and OpenAI SDK compatible agents

Getting Started

Quick Start

Follow the quick start example on our docs site.

Examples

Explore working examples in the examples directory.

MCP Server

The AgentRPC TypeScript SDK includes an optional MCP (Model Context Protocol) server.

ANGENTRPC_API_SECRET=YOUR_API_SECRET npx agentrpc mcp

This launches an MCP-compliant server for external AI models to interact with your registered tools.

Claude Desktop Integration

Add to your claude_desktop_config.json:

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

More Info

Cursor Integration

Add to your ~/.cursor/mcp.json:

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

More Info

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

This repository contains all the open-source components and SDKs for AgentRPC.