RAG

RAG with MCP

python

4

GitHub Stars

python

Language

6 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": {
    "plaban1981-rag-mcp": {
      "command": "python",
      "args": [
        "mcp_server.py"
      ]
    }
  }
}

You will run an MCP server that powers a retrieval-augmented workflow and exposes a streaming interface for interactive queries. This guide shows how to start the server locally, connect with an MCP client, and perform common usage patterns for streaming results and tool access.

How to use

Start the MCP server from your terminal in the project folder by running the Python script. The server will become available at http://localhost:8000. You can connect an MCP client to the SSE endpoint to receive streaming results and to the messages endpoint to send queries. The initial HTTP GET request to the root path may return a 404, while streaming and message endpoints are available once the server is up.

How to install

Prerequisites: ensure you have Python installed on your system. You will run the server from a virtual environment and start it with a Python command.

Create and activate a virtual environment.
Install any required dependencies for the MCP server if you have a requirements file or setup script in your project.
Start the MCP server using the following command.

Notes and troubleshooting

When the server starts, you will see output indicating the server process ID, the startup status, and the URL where the server is listening. For example, you may observe a line similar to “Server will be available at http://localhost:8000” and a follow-up line showing the local UVicorn server is running. The root path may respond with 404, while the SSE endpoint will typically respond with 200 OK. You can then proceed to send messages to the server for processing and receive streaming results via the SSE channel.

Example runtime configuration (local MCP server)

{
  "mcpServers": [
    {
      "name": "mcp_server",
      "type": "stdio",
      "command": "python",
      "args": ["mcp_server.py"]
    }
  ]
}