mcp-local-rag

What is mcp-local-rag?

mcp-local-rag is a local server implementing the Model Context Protocol (MCP) to provide Retrieval-Augmented Generation (RAG) capabilities for large language models. It allows LLMs to perform web searches and incorporate real-time information into their responses, effectively bypassing the need for external APIs for web access.

How to use mcp-local-rag?

mcp-local-rag can be integrated into your MCP server configuration. There are two primary methods for installation:

• Run Directly via uvx: This method requires uv to be installed. You add a specific JSON configuration to your MCP server settings, which instructs it to run mcp-local-rag using uvx with Python 3.10.

  {
    "mcpServers": {
      "mcp-local-rag":{
        "command": "uvx",
          "args": [
            "--python=3.10",
            "--from",
            "git+https://github.com/nkapila6/mcp-local-rag",
            "mcp-local-rag"
          ]
        }
    }
  }
  

• Using Docker (recommended): This method requires Docker to be installed. You add a different JSON configuration to your MCP server settings, which instructs it to run mcp-local-rag as a Docker container.

  {
    "mcpServers": {
      "mcp-local-rag": {
        "command": "docker",
        "args": [
          "run",
          "--rm",
          "-i",
          "--init",
          "-e",
          "DOCKER_CONTAINER=true",
          "ghcr.io/nkapila6/mcp-local-rag:latest"
        ]
      }
    }
  }
  

Once configured, any MCP client that supports tool calling can leverage mcp-local-rag.

Key Features of mcp-local-rag

• Local RAG Implementation: Runs entirely locally, ensuring data privacy and reducing reliance on external services.
• No External APIs: Eliminates the need for third-party web search APIs, simplifying setup and reducing potential costs.
• Google MediaPipe Text Embedder: Utilizes MediaPipe for efficient text embedding.
• DuckDuckGo Search: Leverages DuckDuckGo for web search queries.
• Context Extraction: Extracts relevant context from search results to provide to the LLM.
• Compatibility: Works with any MCP client that supports tool calling, tested with Claude Desktop, Cursor, and Goose.

Use Cases of mcp-local-rag

• Enhancing LLM Knowledge: Provides LLMs with up-to-date information for questions requiring recent web data, such as current events or newly released products.
• Real-time Information Retrieval: Enables LLMs to perform live web searches and incorporate fresh knowledge into their responses.
• Offline/Private Environments: Ideal for scenarios where internet access is restricted or data privacy is paramount, as it runs locally.
• Tool for LLM Development: Serves as a valuable tool for developers building LLM applications that require dynamic web content integration.

FAQ about mcp-local-rag

• What is MCP? MCP stands for Model Context Protocol, a protocol that allows language models to interact with external tools and services.
• Does mcp-local-rag require an internet connection? Yes, it requires an internet connection to perform web searches via DuckDuckGo, but it does not rely on external APIs for its core functionality.
• Which LLMs are compatible with mcp-local-rag? Any LLM that supports the Model Context Protocol and tool calling should be compatible. It has been tested with Claude Desktop, Cursor, and Goose.
• How does mcp-local-rag ensure security? MseeP conducts security audits on every MCP server, and you can view the security audit for mcp-local-rag on the MseeP website.

mcp-local-rag

"primitive" RAG-like web search model context protocol (MCP) server that runs locally. ✨ no APIs ✨

Installation

Locate your MCP config path here or check your MCP client settings.

Run Directly via uvx

This is the easiest and quickest method. You need to install uv for this to work. Add this to your MCP server configuration:

{
  "mcpServers": {
    "mcp-local-rag":{
      "command": "uvx",
        "args": [
          "--python=3.10",
          "--from",
          "git+https://github.com/nkapila6/mcp-local-rag",
          "mcp-local-rag"
        ]
      }
  }
}

Using Docker (recommended)

Ensure you have Docker installed. Add this to your MCP server configuration:

{
  "mcpServers": {
    "mcp-local-rag": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "--init",
        "-e",
        "DOCKER_CONTAINER=true",
        "ghcr.io/nkapila6/mcp-local-rag:latest"
      ]
    }
  }
}

Security audits

MseeP does security audits on every MCP server, you can see the security audit of this MCP server by clicking here.

MCP Clients

The MCP server should work with any MCP client that supports tool calling. Has been tested on the below clients.

• Claude Desktop
• Cursor
• Goose
• Others? You try!

Examples on Claude Desktop

When an LLM (like Claude) is asked a question requiring recent web information, it will trigger mcp-local-rag.

When asked to fetch/lookup/search the web, the model prompts you to use MCP server for the chat.

In the example, have asked it about Google's latest Gemma models released yesterday. This is new info that Claude is not aware about.

Result

mcp-local-rag performs a live web search, extracts context, and sends it back to the model—giving it fresh knowledge:

Contributing

Have ideas or want to improve this project? Issues and pull requests are welcome!

License

This project is licensed under the MIT License.