🚀 Nchan MCP Transport
@ConechoAI
About 🚀 Nchan MCP Transport
The best way to deploy mcp server. A high-performance WebSocket/SSE transport layer & gateway for Anthropic's MCP (Model Context Protocol) — powered by Nginx, Nchan, and FastAPI.
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"nchan-mcp-transport": {
"command": "python",
"args": [
"-m",
"httmcp",
"-f",
"gpt_actions_openapi.json",
"-p",
"http://nchan:80"
]
}
}
}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 🚀 Nchan MCP Transport?
It is a high-performance WebSocket/SSE transport layer and gateway for Anthropic’s MCP (Model Context Protocol), powered by Nginx, Nchan, and FastAPI. It is designed for building real-time, scalable AI integrations with Claude and other LLM agents.
How to use 🚀 Nchan MCP Transport?
Install the server SDK with pip install httmcp, then clone the repository and run docker-compose up -d. Define tools using Python decorators (@server.tool()), optionally expose OpenAPI services with OpenAPIMCP.from_openapi(), or use the CLI for one‑click GPTs Actions to MCP deployment.
Key features of 🚀 Nchan MCP Transport
- Dual protocol support: WebSocket and SSE with automatic detection
- High performance pub/sub built on Nginx + Nchan
- Fully MCP‑compliant transport (JSON‑RPC 2.0)
- Auto‑generate MCP tools from any OpenAPI spec
- Use Python decorators to register tools and resources
- Background task queue with live progress updates via push
- Dockerized deployment ready
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
How is this better than native MCP?
Native MCP (HTTP+SSE) struggles with long tasks, network instability, and high concurrency; WebSockets aren’t natively supported by Claude. This project bridges that gap with Nginx/Nchan‑level scalability and real‑time event delivery.
How do I define a tool?
Use the Python decorator @server.tool() on an async function. Example: async def search_docs(query: str) -> str: return f"Searching for {query}...".
Can I connect existing APIs to Claude?
Yes – use OpenAPIMCP.from_openapi() to expose any OpenAPI spec as MCP tools, or use the CLI for one‑click deployment from GPTs Actions OpenAPI specs.
What are the runtime requirements?
Python 3.9+, Nginx with the Nchan module (pre‑installed in the Docker image), and Docker / Docker Compose.
Does Claude support WebSocket natively?
No, Claude does not natively support WebSocket. This project provides a bridge that allows WebSocket‑based communication with Claude clients.
Frequently asked questions
How is this better than native MCP?
Native MCP (HTTP+SSE) struggles with long tasks, network instability, and high concurrency; WebSockets aren’t natively supported by Claude. This project bridges that gap with Nginx/Nchan‑level scalability and real‑time event delivery.
How do I define a tool?
Use the Python decorator `@server.tool()` on an async function. Example: `async def search_docs(query: str) -> str: return f"Searching for {query}..."`.
Can I connect existing APIs to Claude?
Yes – use `OpenAPIMCP.from_openapi()` to expose any OpenAPI spec as MCP tools, or use the CLI for one‑click deployment from GPTs Actions OpenAPI specs.
What are the runtime requirements?
Python 3.9+, Nginx with the Nchan module (pre‑installed in the Docker image), and Docker / Docker Compose.
Does Claude support WebSocket natively?
No, Claude does not natively support WebSocket. This project provides a bridge that allows WebSocket‑based communication with Claude clients.
Basic information
More Other MCP servers
IDA Pro MCP
mrexodiaAI-powered reverse engineering assistant that bridges IDA Pro with language models through MCP.

Glasswarp
GlasswarpSee and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.
MCP Registry
modelcontextprotocolA community driven registry service for Model Context Protocol (MCP) servers.
Awesome Mlops
visengerA curated list of references for MLOps
ghidraMCP
LaurieWiredMCP Server for Ghidra
Comments