Mcp Librechat
What is Mcp Librechat
mcp-librechat is a setup for creating MCP servers using LibreChat, which allows for the integration of various tools and functionalities into a chat interface.
Use cases
Use cases include creating chatbots that perform calculations, integrating search functionalities into chat applications, and developing interactive AI assistants.
How to use
To use mcp-librechat, clone the repository, configure the environment variables, including API keys, and run the server using Docker Compose. Access the LibreChat interface at https://localhost:3080/ after setup.
Key features
Key features include the ability to run multiple MCP servers, integration with tools like FastMCP for number addition and Brave search engine, and a user-friendly interface for managing agents and tools.
Where to use
mcp-librechat can be used in various fields including software development, AI research, and any application requiring interactive chat interfaces with integrated tools.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Overview
What is Mcp Librechat
mcp-librechat is a setup for creating MCP servers using LibreChat, which allows for the integration of various tools and functionalities into a chat interface.
Use cases
Use cases include creating chatbots that perform calculations, integrating search functionalities into chat applications, and developing interactive AI assistants.
How to use
To use mcp-librechat, clone the repository, configure the environment variables, including API keys, and run the server using Docker Compose. Access the LibreChat interface at https://localhost:3080/ after setup.
Key features
Key features include the ability to run multiple MCP servers, integration with tools like FastMCP for number addition and Brave search engine, and a user-friendly interface for managing agents and tools.
Where to use
mcp-librechat can be used in various fields including software development, AI research, and any application requiring interactive chat interfaces with integrated tools.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Content
LibreChat MCP server setup
This is a small setup that demonstrates how you can set up MCP servers with LibreChat
Overview
This repo contains a docker-compose.yml and librechat.yaml that configures LibreChat with a few different MCP servers (in the servers/ directory), in this case two simple ones (at the time of writing):
- A python/FastMCP based server that exposes a single tool to add numbers (note that this style of MCP server declaration will be in the official MCP python sdk soon)
- A typescript based server that exposes a tool that allows the LLM to use the brave search engine. This requires an API key for brave. This is copied from the official servers, so check these docs for API setup
Requirements
You’ll need docker, with docker compose. If you’re on MacOS or Windows you’ll probably need something like docker desktop.
Setup
Make a copy of the .env.example file:
cp .env.example .env
And then add in your Anthropic api key (or OpenAI, but I haven’t tested this out). And also configure the BRAVE_API_KEY for the servers/brave-search MCP server.
Once this is done, to start the server you should be able to run:
docker compose up -d --build
Note: you might need to use sudo
Then you should be able to go to https://localhost:3080/ to see LibreChat. Once you’ve signed up and made an account you should be able to configure an agent, and “add tools”. On the second page of tools you should see our MCP tools! Save the agent and start a conversation with it to try out the MCP tools.
Updating the MCP servers
You should be able to now add new tools to the existing MCP servers (or add whole new servers, but for that you’ll need to update the docker-compose.yml and librechat.yaml) re-start the docker compose (docker compose down && docker compose up -d --build) and your changes should be present.
Dev Tools Supporting MCP
The following are the main code editors that support the Model Context Protocol. Click the link to visit the official website for more information.
