{"type":"mcp_server","name":"agentrpc","description":"AgentRPC is a universal RPC layer for connecting AI agents to functions across networks and languages, supporting multi-language tools, private networks, and offering observability and automatic failover. AgentRPC is a","category":"AI","language":"TypeScript","stars":132,"forks":0,"owner":"agentrpc","github_url":"https://github.com/agentrpc/agentrpc","homepage":"https://www.agentrpc.com","setup":"## Setup\n\n1.  Follow the [quick start](https://docs.agentrpc.com/quickstart) example on our docs site.\n2.  Explore working examples in the [examples](./examples) directory.\n3.  Launch an MCP-compliant server using: `ANGENTRPC_API_SECRET=YOUR_API_SECRET npx agentrpc mcp`.\n4.  For Claude Desktop Integration, add the provided JSON configuration to your `claude_desktop_config.json`.\n5.  For Cursor Integration, add the provided JSON configuration to your `~/.cursor/mcp.json`.","tools":"## Available Tools\n\n1.  Multi-language Support (Connect to tools in TypeScript, Go, Python and .NET (coming soon))\n2.  Private Network Support (Register functions in private VPCs with no open ports required)\n3.  Long-running Functions (Long polling SDKs allow function calls beyond HTTP timeout limits)\n4.  Full Observability (Comprehensive tracing, metrics, and events for complete visibility)\n5.  Automatic Failover (Intelligent health tracking with automatic failover and retries)\n6.  Framework Compatibility (Out-of-the-box support for MCP and OpenAI SDK compatible agents)\n7.  Model Context Protocol (MCP) server (Launches an MCP-compliant server for external AI models to interact with your registered tools)","faq":null,"created_at":"2025-03-17T03:27:52+00:00","updated_at":"2025-03-27T08:44:09+00:00","source_url":"https://model-context-protocol.com/servers/ai-agent-universal-rpc-layer","related_articles":[]}