mcp-unify
Provides tools for interacting with GitHub's API, enabling AI agents to manage repositories, issues, pull requests, and other GitHub resources.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@mcp-unifylist my GitHub issues"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
mcp-unify
Unify multiple MCP servers behind a single endpoint. Lazy loading, auto-cleanup, Python plugins, role-based filtering.
The Problem
If you use Claude Code or any MCP client with 3+ servers, you get:
Multiple subprocesses (~50MB each)
Duplicated config
No centralized control
No way to add custom Python tools without a full MCP server
Related MCP server: mcp-uni
The Solution
mcp-unify runs one process that proxies N MCP servers on-demand:
Claude Code / MCP Client
│
▼
mcp-unify (1 process)
├─ [plugin] Python @tool functions ← in-process, 0 overhead
├─ [lazy] filesystem-server ← subprocess spawned on first call
├─ [lazy] github-server ← subprocess spawned on first call
└─ [lazy] playwright ← subprocess spawned on first call
↑
5 min idle → auto-killInstall
pip install mcp-unifyQuick Start
1. Create gateway.yaml
servers:
filesystem:
command: npx
args: ["-y", "@modelcontextprotocol/server-filesystem", "/tmp"]
github:
command: npx
args: ["-y", "@modelcontextprotocol/server-github"]
env:
GITHUB_TOKEN: "${GITHUB_TOKEN}"
enabled: false
idle_timeout: 3002. Run
stdio (for Claude Code):
mcp-unify stdio --config gateway.yamlSSE (for remote clients):
mcp-unify serve --config gateway.yaml --port 87653. Add to Claude Code
{
"mcpServers": {
"gateway": {
"command": "mcp-unify",
"args": ["stdio", "--config", "/path/to/gateway.yaml"]
}
}
}Python Plugins
Add custom tools with zero boilerplate:
from mcp_gateway import tool
@tool(description="Add two numbers")
def add(a: float, b: float) -> float:
return a + bReference in your config:
plugins:
- module: my_tools.calculatorOr register programmatically:
from mcp_gateway import MCPGateway
gw = MCPGateway()
gw.register_tool(add)
await gw.serve_stdio()Sync functions run via asyncio.to_thread(). Async functions run natively. Schemas are auto-generated from type hints.
Role-Based Filtering
Control which tools each client sees:
roles:
admin: null # all tools
readonly: ["filesystem_*", "gateway_status"]
developer: ["github_*", "filesystem_*", "gateway_*"]Set the role via environment variable:
MCP_GATEWAY_ROLE=readonly mcp-unify stdioUses glob patterns — filesystem_* matches all tools from the filesystem server.
How It Works
Lazy Loading: Servers aren't started until a tool is called. Each server gets a
_<name>_connectplaceholder tool. Calling it spawns the subprocess and discovers real tool schemas viasession.list_tools().Auto-Cleanup: A background task checks every 60s and kills server subprocesses idle for longer than
idle_timeout(default: 5 min).Real Schemas: Tool schemas are discovered from the actual server via the MCP SDK, not hardcoded. You always get accurate
inputSchema.Plugin Tools: Python functions decorated with
@toolrun in the gateway process. Type hints are introspected to generate JSON Schema automatically.
API
from mcp_gateway import MCPGateway, tool
# From YAML
gw = MCPGateway.from_config("gateway.yaml")
# Programmatic
gw = MCPGateway()
gw.register_tool(my_function, name="my_tool", description="...")
# Serve
await gw.serve_stdio() # stdio mode
await gw.serve_sse() # SSE mode on :8765
# Context manager
async with MCPGateway() as gw:
await gw.serve_stdio()CLI
mcp-unify serve [--config FILE] [--host HOST] [--port PORT]
mcp-unify stdio [--config FILE]
mcp-unify list [--config FILE]Config discovery: --config > ./gateway.yaml > ./mcp-unify.yaml
Requirements
Python >= 3.10
mcp >= 1.0.0starlette >= 0.27.0uvicorn >= 0.23.0pyyaml >= 6.0
License
MIT
This server cannot be deployed
Maintenance
Related MCP Connectors
Trust, freshness, policy, and discovery layer for public MCP servers.
The MCP server that finds MCP servers. Aggregates Official Registry, Glama, and Smithery.
Governed MCP gateway: one endpoint for your tools, with credential custody and audit log.
Billing proxy for MCP servers. Adds Stripe and x402 crypto payments without writing billing code.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceProxy that aggregates multiple MCP servers and presents them as a unified interface, allowing clients to access resources from multiple servers transparently.4-
- AlicenseNot gradedqualityCmaintenanceA universal MCP server that acts as a unified gateway for dynamically connecting and managing multiple MCP servers via a single HTTP endpoint.7 npm6MIT
- AlicenseNot gradedqualityCmaintenanceAggregates multiple backend MCP servers into a single unified interface with optional web management UI for tool control and configuration.60 npm193MIT
- AlicenseNot gradedqualityAmaintenanceAggregates multiple MCP servers into a single HTTP endpoint with tool namespacing, dashboard, and REST API for management.13 npmMIT