MCP-Bridge
A middleware to provide an openAI compatible endpoint that can call MCP tools
MCP-Bridge is a middleware server that provides an OpenAI-compatible API endpoint for accessing Model Context Protocol (MCP) tools. It acts as a bridge between OpenAI API clients and MCP servers, enabling any OpenAI-compatible application to leverage MCP tools without native MCP support. The server translates incoming OpenAI API requests, injects MCP tool definitions, forwards them to an inference engine, manages tool calls to MCP servers, and returns formatted responses. It also offers an SSE endpoint for clients with native MCP support and a REST API for direct interaction with MCP primitives.
Key Features
Use Cases
- 01Enable Open Web UI or other OpenAI-compatible chat interfaces to use MCP tools like fetch, filesystem, or database connectors
- 02Aggregate multiple MCP servers behind a single OpenAI API endpoint for unified tool access
- 03Use MCP tools with inference engines like vLLM or Ollama that support tool calling
- 04Connect MCP-native clients like Claude Desktop to multiple MCP servers through the SSE bridge
- 05Test and debug MCP server configurations using the REST API endpoints
- 06Secure MCP tool access with API key authentication in multi-user environments
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
MCP-Bridge — FAQ
What is MCP-Bridge?+
MCP-Bridge is middleware that sits between OpenAI API clients and MCP servers, translating OpenAI API requests to use MCP tools. It also provides an SSE endpoint for MCP-native clients to connect to multiple MCP servers at once.
How do I install MCP-Bridge?+
The recommended installation method is Docker using the provided compose.yml file with a config.json that defines your inference server and MCP servers. You can also install manually using uv sync and running the Python application directly.
What clients work with MCP-Bridge?+
Any OpenAI API-compatible client can use the OpenAI endpoint, including Open Web UI, while MCP-native clients like Claude Desktop can connect via the SSE bridge endpoint. An inference engine with tool call support (vLLM, Ollama) is required.
Do I need API keys to use MCP-Bridge?+
You need an inference server (like vLLM or Ollama) with its base URL configured. Optionally, you can enable API key authentication on MCP-Bridge itself to secure access to your server.
Is MCP-Bridge free and open source?+
Yes, MCP-Bridge is open source and licensed under the MIT License, making it free to use and modify.
What are the prerequisites for running MCP-Bridge?+
You need an inference engine with tool call support (such as vLLM or Ollama), MCP servers you want to expose, and either Docker or Python with uv for manual installation. The MCP servers themselves may have additional requirements.
How do I install MCP-Bridge?+
Open the source repository on GitHub and follow its README. MCP-Bridge is a mcp server — MCP Agents Market links you directly to the official repo.
Is MCP-Bridge free?+
MCP-Bridge is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.