RAGFlow

Provides a complete MCP server for semantic retrieval, datasets, documents, chunks, chats, sessions, and graph-based knowledge management with RAGFlow.

python

0

GitHub Stars

python

Language

4 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": {
    "migoxv-ragflow-mcp": {
      "command": "python",
      "args": [
        "-m",
        "ragflow_mcp.server"
      ],
      "env": {
        "RAGFLOW_URL": "http://localhost:9380/api/v1",
        "RAGFLOW_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

You can run the RAGFlow MCP Server to expose full MCP access for semantic retrieval and knowledge base management against your RAGFlow instance. This server lets you manage datasets, documents, chunks, chats, sessions, and graph-based structures, all through a dedicated MCP interface.

How to use

You use an MCP client to connect to the RAGFlow MCP Server and perform operations such as searching datasets semantically, uploading documents, managing chunks, or building knowledge graphs. Start the server, configure your client with the correct API key and base URL, and then call the available tools to perform tasks like listing datasets, uploading documents, parsing content, or building a knowledge graph. All actions follow the MCP protocol and are designed to be reliable and auditable.

How to install

Prerequisites you need before installation:

Python 3.10+
A running RAGFlow server (v0.16.0+ for core features)
A RAGFlow API key

Install from source by cloning the repository, installing the package in editable mode, and preparing for execution.

git clone https://github.com/Juxsta/ragflow-mcp.git
cd ragflow-mcp
pip install -e .

Run and configure the MCP server

Run the MCP server using Python as a standard stdio server. You will provide your RAGFlow API key and the base API URL via environment variables.

# Start the MCP server (example; adapt paths as needed)
python -m ragflow_mcp.server

Configure Claude Code to use the MCP server

Add a new MCP target named ragflow in your Claude Code settings by providing the API key and the MCP URL, and then run the server module.

claude mcp add ragflow -e RAGFLOW_API_KEY=your-api-key -e RAGFLOW_URL=http://localhost:9380/api/v1 -- python -m ragflow_mcp.server

Alternative local configuration (settings.json)

{
  "mcpServers": {
    "ragflow": {
      "command": "python",
      "args": ["-m", "ragflow_mcp.server"],
      "cwd": "/path/to/ragflow-mcp",
      "env": {
        "RAGFLOW_API_KEY": "your-api-key",
        "RAGFLOW_URL": "http://localhost:9380/api/v1"
      }
    }
  }
}

Available tools

Retrieval

Semantic search across datasets using natural language queries.

Dataset management

Create, list, update, and delete datasets.

Document management

Upload, parse, list, download, and delete documents.

Chunk management

Add, list, update, and delete document chunks.

Chat & sessions

Create and manage chat assistants and chat sessions.

GraphRAG & RAPTOR

Build and query knowledge graphs when supported by the RAGFlow instance.