MCP Bridge API

v1.0.0APIsstable

MCP Bridge is a lightweight, fast, and LLM-agnostic proxy for connecting to multiple Model Context Protocol (MCP) servers through a unified REST API. It enables secure tool execution across diverse environments like mobile, web, and edge devices. Des

mcp-bridge-apimcpai-integration
Share:
64
Stars
0
Downloads
0
Weekly
0/5

What is MCP Bridge API?

MCP Bridge API is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to mcp bridge is a lightweight, fast, and llm-agnostic proxy for connecting to multiple model context protocol (mcp) servers through a unified rest api. it enables secure tool execution across diverse en...

MCP Bridge is a lightweight, fast, and LLM-agnostic proxy for connecting to multiple Model Context Protocol (MCP) servers through a unified REST API. It enables secure tool execution across diverse environments like mobile, web, and edge devices. Des

This server falls under the APIs and Developer Tools categories on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • MCP Bridge is a lightweight, fast, and LLM-agnostic proxy fo

Use Cases

Connect to multiple MCP servers via unified REST API.
Execute tools across mobile, web, and edge devices.
Provide LLM-agnostic proxy for MCP access.
INQUIRELAB

Maintainer

LicenseMIT
Languagetypescript
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx mcp-bridge-api

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use MCP Bridge API

MCP Bridge API is a lightweight, LLM-agnostic REST proxy that exposes multiple Model Context Protocol servers behind a single HTTP endpoint, making MCP tools accessible from environments where stdio transport is impractical — including mobile apps, web frontends, and edge devices. It handles connection management, risk-based execution controls, and optional Docker isolation, letting any HTTP client (not just Claude Desktop) call MCP tools through standard REST calls. Teams building React Native apps, web dashboards, or Python agents that need MCP tool access without embedding an MCP client can use this bridge as a central gateway.

Prerequisites

  • Node.js 18 or higher for the bridge server
  • Python 3.8 or higher if using the bundled Python agent
  • Docker (optional, required only for high-risk tool isolation)
  • Any MCP-compatible client or HTTP client to call the REST API
  • The MCP servers you want to expose already installed on the same machine
1

Clone the repository

Clone the mcp-bridge-api repository to your machine and enter the project directory.

git clone https://github.com/INQUIRELAB/mcp-bridge-api.git
cd mcp-bridge-api
2

Install Node.js dependencies

Install the required Express-based packages for the bridge server.

npm install express cors morgan uuid
3

Configure mcp_config.json

Create or edit mcp_config.json in the project root to declare which MCP servers the bridge should manage. Each entry specifies the command, arguments, optional environment variables, and a riskLevel (1=low/direct, 2=medium/confirm, 3=high/Docker).

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "/tmp"],
      "riskLevel": 2
    },
    "slack": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-slack"],
      "env": {
        "SLACK_BOT_TOKEN": "xoxb-your-token",
        "SLACK_TEAM_ID": "T01234567"
      },
      "riskLevel": 1
    }
  }
}
4

Start the bridge server

Launch the MCP Bridge API. By default it listens on port 3000.

node mcp-bridge.js
5

Verify the bridge is running

Check the health endpoint to confirm the server started and all configured MCP servers are connected.

curl http://localhost:3000/health
6

Call MCP tools via REST

Execute a tool on a connected MCP server using a POST request. Replace 'filesystem' and 'read_file' with the desired server ID and tool name.

curl -X POST http://localhost:3000/servers/filesystem/tools/read_file \
  -H 'Content-Type: application/json' \
  -d '{"path": "/tmp/example.txt"}'

MCP Bridge API Examples

Client configuration

Example mcp_config.json declaring two MCP servers with different risk levels for the bridge to manage.

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "/tmp"],
      "riskLevel": 2
    },
    "slack": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-slack"],
      "env": {
        "SLACK_BOT_TOKEN": "xoxb-your-token",
        "SLACK_TEAM_ID": "T01234567"
      },
      "riskLevel": 1
    }
  }
}

Prompts to try

REST API calls and Python agent commands that demonstrate bridge capabilities.

- GET http://localhost:3000/servers  — list all connected MCP servers
- GET http://localhost:3000/servers/filesystem/tools  — list tools on the filesystem server
- POST http://localhost:3000/servers/filesystem/tools/list_directory {"path":"/tmp"}  — run a tool
- GET http://localhost:3000/health  — check bridge status
- python llm_test.py --mcp-url http://localhost:3000  — run the Python agent against the bridge

Troubleshooting MCP Bridge API

A configured MCP server fails to start when the bridge launches

Run the MCP server command manually (e.g. 'npx -y @modelcontextprotocol/server-filesystem /tmp') to see its error output. Common causes are missing packages or incorrect paths in mcp_config.json.

POST to a tool endpoint returns 'confirmation required'

The tool's riskLevel is set to 2 (medium). Send a POST to /confirmations/{confirmationId} with the confirmation payload returned in the original response to proceed with execution.

Port 3000 is already in use

Pass a custom port when starting: 'node mcp-bridge.js --port 3001', or update the Python agent with '--mcp-port 3001' to match.

Frequently Asked Questions about MCP Bridge API

What is MCP Bridge API?

MCP Bridge API is a Model Context Protocol (MCP) server that mcp bridge is a lightweight, fast, and llm-agnostic proxy for connecting to multiple model context protocol (mcp) servers through a unified rest api. it enables secure tool execution across diverse environments like mobile, web, and edge devices. des It connects AI assistants to external tools and data sources through a standardized interface.

How do I install MCP Bridge API?

Follow the installation instructions on the MCP Bridge API GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with MCP Bridge API?

MCP Bridge API works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is MCP Bridge API free to use?

Yes, MCP Bridge API is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More APIs MCP Servers

Explore all apis servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "mcp-bridge-api": { "command": "npx", "args": ["-y", "mcp-bridge-api"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use MCP Bridge API?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides