OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
MCP & Tooling

MCP Bridge

MCP-Bridge is a Python middleware server that exposes configured Model Context Protocol (MCP) tools through an OpenAI-compatible API and provides an SSE bridge for external MCP clients.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

MCP Bridge — OpenViST project profile in MCP & Tooling.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when an OpenAI-compatible application needs access to MCP tools through middleware, or when an external MCP-aware client needs the bridge's SSE interface.

Who it is for

  • Developers connecting MCP tool servers to applications that use an OpenAI-compatible API.
  • Operators who want to expose configured MCP tools to external clients through an SSE bridge.

Documented uses

  • Let an OpenAI-compatible client invoke configured MCP tools.
  • Configure an inference server and MCP servers behind a common API-facing service.
  • Expose configured MCP tools to MCP-aware external clients over SSE.
  • List the MCP tools available through the running service.
Before you choose

Requirements & limits

The repository is public and lists a MIT license. The project documents Docker and manual installation; using it requires operating the server and configuring an inference engine and MCP servers.

  • The service requires an inference engine with tool-call support; the README reports testing with vLLM and says Ollama should also be compatible.
  • The README recommends Docker, while manual installation uses uv sync and uv run mcp_bridge/main.py.
  • API-key authentication can be enabled by configuration; the README says authentication is skipped if the API-key field is empty or absent.
  • The repository calls the project soft deprecated because Open WebUI supports MCP natively from version 0.6.31.
  • MCP resources are planned, and the README's feature list distinguishes streaming chat completions with MCP from streaming completions it says are not implemented.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up MCP-Bridge manually

Prepare a manual installation with an inference server and one configured MCP server.

Before you start

  • A cloned MCP-Bridge repository
  • An inference engine with tool-call support and its OpenAI-compatible base URL
  • The MCP server command and arguments to place in the configuration

Steps

  1. Install project dependencies with the command below.

    Documented command
    uv sync

  2. Create config.json in the repository root with the inference server base URL and API key, plus the MCP server command and arguments.

  3. Start the application with the command below.

    Documented command
    uv run mcp_bridge/main.py

Expected outcome

The README says the running service can be used through the OpenAI API; its documentation page includes an endpoint for listing configured MCP tools.

Guide limitations

  • The inference engine must support tool calls.
  • MCP resources are listed as planned, not supported.
  • The project is described as soft deprecated because Open WebUI supports MCP natively from version 0.6.31.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
5 November 2026
Original website title
SecretiveShell/MCP-Bridge · Official homepage title · retained in its source language
How we review sourcesSuggest a correction

Related options

Open WebUI native MCP support — The README says Open WebUI supports MCP natively from version 0.6.31 and cites this as the reason MCP-Bridge should be considered soft deprecated.