AgentRPC

Name: AgentRPC
Author: agentrpc

@agentrpc

Repository Homepage

About AgentRPC

A universal RPC layer for AI agents. Connect to any function, any language, any framework, in minutes.

#agentrpc #rpc-layer #ai-agents

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "agentrpc": {
      "command": "npx",
      "args": [
        "agentrpc",
        "mcp"
      ],
      "env": {
        "ANGENTRPC_API_SECRET": "YOUR_API_SECRET"
      }
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is AgentRPC?

AgentRPC is a universal RPC layer for AI agents that allows connecting to any function, in any language, across network boundaries. It wraps functions in a universal RPC interface and connects them to a hosted RPC server accessible through the Model Context Protocol (MCP) and OpenAI-compatible tool definitions. It is ideal for services deployed in private VPCs, Kubernetes clusters, or multiple cloud environments.

How to use AgentRPC?

Install the AgentRPC SDK for your language (TypeScript, Go, Python, .NET coming soon), register your functions, then launch the MCP server with AGENTRPC_API_SECRET=YOUR_API_SECRET npx agentrpc mcp. Integrate with Claude Desktop or Cursor by adding the server configuration to their respective JSON config files.

Key features of AgentRPC

Multi-language support (TypeScript, Go, Python, .NET coming soon)
Private network support with no open ports required
Long-running functions via long polling SDKs
Full observability with tracing, metrics, and events
Automatic failover and intelligent health tracking
Framework compatibility with MCP and OpenAI SDK agents

Use cases of AgentRPC

Connect AI agents to services running in private VPCs
Orchestrate functions across Kubernetes clusters
Integrate tools from multiple cloud environments
Execute long-running functions beyond HTTP timeout limits

FAQ from AgentRPC

What languages does AgentRPC support?

AgentRPC supports TypeScript, Go, and Python, with .NET support coming soon.

How does AgentRPC handle private networks?

AgentRPC uses long polling SDKs that require no open ports, allowing functions in private VPCs to be registered and called securely.

What protocols does AgentRPC support?

AgentRPC supports the Model Context Protocol (MCP) and OpenAI-compatible tool definitions, making it compatible with OpenAI, Anthropic, LiteLLM, OpenRouter, and similar agents.

How do I get started with AgentRPC?

Follow the quick start guide on the AgentRPC documentation site (docs.agentrpc.com/quickstart) and explore working examples in the examples directory.