mcpserver.so LogoMCPFly
Submit
Nchan MCP Transport logo

Nchan MCP Transport

by ConechoAI

AI/Integrationmcp transportwebsocketsseanthropicnginxfastapiclaudeopenapi

A high-performance WebSocket/SSE transport layer & gateway for Anthropic's MCP, powered by Nginx, Nchan, and FastAPI. It's designed for building real-time, scalable AI integrations with Claude and other LLM agents.

View on GitHub

Last updated: N/A

What is Nchan MCP Transport?

Nchan MCP Transport provides a real-time API gateway for MCP clients (like Claude) to communicate with tools and services over WebSocket or SSE. It leverages Nginx and Nchan for low-latency pub/sub and integrates with FastAPI for backend logic and OpenAPI tooling.

How to use Nchan MCP Transport?

To use Nchan MCP Transport, first install the server SDK using pip install httmcp. Then, run the demo using Docker Compose. Define your tools using Python decorators, and optionally expose an OpenAPI service. Finally, you can use the HTTMCP CLI for one-click deployment of GPTs Actions to MCP servers.

Key features of Nchan MCP Transport

  • Dual Protocol Support (WebSocket and SSE)

  • High Performance Pub/Sub (Nginx + Nchan)

  • MCP-Compliant Transport (JSON-RPC 2.0)

  • OpenAPI Integration

  • Tool / Resource System

  • Asynchronous Execution

  • Dockerized Deployment

Use cases of Nchan MCP Transport

  • Claude plugin server over WebSocket/SSE

  • Real-time LLM agent backend (LangChain/AutoGen style)

  • Connect Claude to internal APIs (via OpenAPI)

  • High-performance tool/service bridge for MCP

FAQ from Nchan MCP Transport

Why use Nchan MCP Transport?

It offers web-scale performance, a FastAPI-powered backend for tools, real-time event delivery to Claude clients, and plug-and-play OpenAPI to Claude integration.

What are the requirements?

You need Nginx with the Nchan module, Python 3.9+, and Docker/Docker Compose.

What is MCP?

MCP (Model Context Protocol) lets AI assistants like Claude talk to external tools.

What is HTTMCP SDK?

HTTMCP SDK is a full MCP protocol implementation.

How does it handle concurrency?

It uses Nginx + Nchan to handle thousands of concurrent connections with low latency.