RAGFlow

Provides an MCP server to search knowledge bases and chat with an AI assistant using datasets.

other

0

GitHub Stars

other

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "aitech-team-ragflow-mcp-server-continue": {
      "command": "uvx",
      "args": [
        "ragflow-mcp-server",
        "--api-key",
        "ragflow-dhMzViYzJlMTM1NjExZjBiNWU5MDI0Mm",
        "--base-url",
        "http://172.16.33.66:8060"
      ]
    }
  }
}

You run the ragflow MCP Server to search knowledge bases and chat with an AI assistant. It exposes endpoints to list datasets, create chat assistants, chat with them, and retrieve relevant knowledge from your datasets. This guide shows practical steps to install, configure, and use the server with MCP clients.

How to use

You interact with the MCP server through an MCP client. Start by selecting a server configuration, then perform the following typical workflows: - List datasets to see available knowledge bases. - Create a chat assistant by choosing a dataset. - Open a chat session with that assistant and ask questions. - Retrieve information by querying against one or more datasets. These actions enable you to build a chat-powered assistant that can query your knowledge repository and provide answers.

How to install

Prerequisites: You need a runtime that can launch MCP servers via the MCP protocol. In practical terms, you will use the UV runtime (uv or uvx) to run the MCP server executable.

Choose one of the following startup configurations depending on your environment. The server is started as a standalone process that communicates over stdio.

{
  "servers": {
    "ragflow-mcp-server": {
      "command": "uvx",
      "args": [
        "ragflow-mcp-server",
        "--api-key=ragflow-dhMzViYzJlMTM1NjExZjBiNWU5MDI0Mm",
        "--base-url=http://172.16.33.66:8060"
      ]
    }
  }
}

If you use the Continue configuration, add a YAML block with the same startup intent in your configuration file.

mcpServers:
  - name: RAGFlow Server
    command: uvx
    args:
      - ragflow-mcp-server
      - --api-key
      - ragflow-dhMzViYzJlMTM1NjExZjBiNWU5MDI0Mm
      - --base-url
      - http://172.16.33.66:8060

For a published server setup, you can run the server with the following minimal command through the MCP runtime if you prefer the published configuration.

{
  "type": "stdio",
  "name": "ragflow_mcp_continue",
  "command": "uvx",
  "args": [
    "ragflow-mcp-server-continue"
  ]
}

Available tools

list_datasets

Lists all datasets and returns their IDs and names.

create_chat

Creates a new chat assistant by providing a name and a dataset ID; returns the assistant ID, name, and session ID.

chat

Interacts with a chat assistant by sending a session ID and a question to receive an answer.

retrieve

Retrieves information relevant to a question by querying specified datasets and returns matched content.