ollama-mcp-bridge
Bridge between Ollama and MCP servers, enabling local LLMs to use Model Context Protocol tools
The ollama-mcp-bridge is a TypeScript-based connector that allows local language models running through Ollama to interact with Model Context Protocol servers. It translates between Ollama's output format and MCP's JSON-RPC protocol, enabling open-source models like Qwen 2.5 to access the same tool ecosystem available to Claude. The bridge supports multiple simultaneous MCP server connections, intelligent tool routing, and structured output validation, bringing enterprise-grade capabilities to locally-hosted AI assistants.
Key Features
Use Cases
- 01Running a fully local AI assistant with filesystem access, web search, and email capabilities
- 02Enabling open-source models to manipulate files and directories through natural language commands
- 03Building research workflows that combine web search, GitHub interactions, and document management
- 04Creating AI agents that generate images via Flux while managing Gmail and Google Drive
- 05Developing privacy-focused assistants that keep all data and processing on local hardware
- 06Prototyping MCP tool integrations with smaller, faster models before deploying to production
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.
ollama-mcp-bridge — FAQ
What is ollama-mcp-bridge?+
The ollama-mcp-bridge is a TypeScript application that connects local language models running in Ollama to Model Context Protocol servers, allowing open-source models to use tools like filesystem operations, web search, GitHub integration, and email management. It acts as a translation layer between Ollama's output format and the MCP JSON-RPC protocol.
How do I install and set up ollama-mcp-bridge?+
First install Ollama and pull a compatible model like qwen2.5-coder:7b-instruct. Then install the desired MCP servers globally via npm, configure API keys for services like Brave Search and GitHub, and edit bridge_config.json to define your MCP server paths and LLM settings. Finally, run npm run start to launch the bridge.
Which AI clients work with ollama-mcp-bridge?+
This bridge is a standalone application that works with Ollama as the LLM backend and connects to standard MCP servers. It runs independently as a command-line interface rather than integrating into existing AI clients like Claude Desktop or Cursor.
What API keys and prerequisites are needed?+
You need Ollama installed locally with a pulled model, Node.js for running the bridge, and API keys for optional services: BRAVE_API_KEY for web search, GITHUB_PERSONAL_ACCESS_TOKEN for GitHub access, and REPLICATE_API_TOKEN for Flux image generation. Gmail and Google Drive require OAuth authentication via the MCP server's auth command.
Is ollama-mcp-bridge free and open source?+
Yes, the bridge itself is open source and available on GitHub. However, some integrated services like Brave Search API, Replicate (for Flux), and GitHub may have their own usage limits or pricing for API access.
Can I use other Ollama models besides Qwen?+
Yes, the bridge is designed to work with any Ollama-compatible model. The default configuration uses Qwen 2.5 7B, but you can change the model in bridge_config.json to use other models that support structured output and tool calling capabilities.
How do I install ollama-mcp-bridge?+
Open the source repository on GitHub and follow its README. ollama-mcp-bridge is a mcp server — MCP Agents Market links you directly to the official repo.
Is ollama-mcp-bridge free?+
ollama-mcp-bridge is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.