Mcp Local Rag

io.github.damoqiongqiu/mcp-local-rag · v0.21.0

Semantic code & doc search with keyword boost. AST code nav, auto HF mirror, local, privacy-first.

Source code Set it up

Connect Runs on your machine

Terminal

claude mcp add mcp-local-rag -e BASE_DIR=<BASE_DIR> -e BASE_DIRS=<BASE_DIRS> -e DB_PATH=<DB_PATH> -e CACHE_DIR=<CACHE_DIR> -e MODEL_NAME=<MODEL_NAME> -e MAX_FILE_SIZE=<MAX_FILE_SIZE> -e RAG_MAX_DISTANCE=<RAG_MAX_DISTANCE> -e RAG_GROUPING=<RAG_GROUPING> -e RAG_MAX_FILES=<RAG_MAX_FILES> -e CHUNK_MIN_LENGTH=<CHUNK_MIN_LENGTH> -e RAG_DEVICE=<RAG_DEVICE> -e RAG_DTYPE=<RAG_DTYPE> -e RAG_HYBRID_WEIGHT=<RAG_HYBRID_WEIGHT> -e RAG_WATCH=<RAG_WATCH> -e HF_ENDPOINT=<HF_ENDPOINT> -e HF_AUTO_MIRROR=<HF_AUTO_MIRROR> -e HTTPS_PROXY=<HTTPS_PROXY> -e HTTP_PROXY=<HTTP_PROXY> -- npx -y @damoqiongqiu/[email protected]

claude_desktop_config.json

{
  "mcpServers": {
    "mcp-local-rag": {
      "command": "npx",
      "args": [
        "-y",
        "@damoqiongqiu/[email protected]"
      ],
      "env": {
        "BASE_DIR": "<BASE_DIR>",
        "BASE_DIRS": "<BASE_DIRS>",
        "DB_PATH": "<DB_PATH>",
        "CACHE_DIR": "<CACHE_DIR>",
        "MODEL_NAME": "<MODEL_NAME>",
        "MAX_FILE_SIZE": "<MAX_FILE_SIZE>",
        "RAG_MAX_DISTANCE": "<RAG_MAX_DISTANCE>",
        "RAG_GROUPING": "<RAG_GROUPING>",
        "RAG_MAX_FILES": "<RAG_MAX_FILES>",
        "CHUNK_MIN_LENGTH": "<CHUNK_MIN_LENGTH>",
        "RAG_DEVICE": "<RAG_DEVICE>",
        "RAG_DTYPE": "<RAG_DTYPE>",
        "RAG_HYBRID_WEIGHT": "<RAG_HYBRID_WEIGHT>",
        "RAG_WATCH": "<RAG_WATCH>",
        "HF_ENDPOINT": "<HF_ENDPOINT>",
        "HF_AUTO_MIRROR": "<HF_AUTO_MIRROR>",
        "HTTPS_PROXY": "<HTTPS_PROXY>",
        "HTTP_PROXY": "<HTTP_PROXY>"
      }
    }
  }
}

~/.cursor/mcp.json

{
  "mcpServers": {
    "mcp-local-rag": {
      "command": "npx",
      "args": [
        "-y",
        "@damoqiongqiu/[email protected]"
      ],
      "env": {
        "BASE_DIR": "<BASE_DIR>",
        "BASE_DIRS": "<BASE_DIRS>",
        "DB_PATH": "<DB_PATH>",
        "CACHE_DIR": "<CACHE_DIR>",
        "MODEL_NAME": "<MODEL_NAME>",
        "MAX_FILE_SIZE": "<MAX_FILE_SIZE>",
        "RAG_MAX_DISTANCE": "<RAG_MAX_DISTANCE>",
        "RAG_GROUPING": "<RAG_GROUPING>",
        "RAG_MAX_FILES": "<RAG_MAX_FILES>",
        "CHUNK_MIN_LENGTH": "<CHUNK_MIN_LENGTH>",
        "RAG_DEVICE": "<RAG_DEVICE>",
        "RAG_DTYPE": "<RAG_DTYPE>",
        "RAG_HYBRID_WEIGHT": "<RAG_HYBRID_WEIGHT>",
        "RAG_WATCH": "<RAG_WATCH>",
        "HF_ENDPOINT": "<HF_ENDPOINT>",
        "HF_AUTO_MIRROR": "<HF_AUTO_MIRROR>",
        "HTTPS_PROXY": "<HTTPS_PROXY>",
        "HTTP_PROXY": "<HTTP_PROXY>"
      }
    }
  }
}

.vscode/mcp.json

{
  "servers": {
    "mcp-local-rag": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@damoqiongqiu/[email protected]"
      ],
      "env": {
        "BASE_DIR": "<BASE_DIR>",
        "BASE_DIRS": "<BASE_DIRS>",
        "DB_PATH": "<DB_PATH>",
        "CACHE_DIR": "<CACHE_DIR>",
        "MODEL_NAME": "<MODEL_NAME>",
        "MAX_FILE_SIZE": "<MAX_FILE_SIZE>",
        "RAG_MAX_DISTANCE": "<RAG_MAX_DISTANCE>",
        "RAG_GROUPING": "<RAG_GROUPING>",
        "RAG_MAX_FILES": "<RAG_MAX_FILES>",
        "CHUNK_MIN_LENGTH": "<CHUNK_MIN_LENGTH>",
        "RAG_DEVICE": "<RAG_DEVICE>",
        "RAG_DTYPE": "<RAG_DTYPE>",
        "RAG_HYBRID_WEIGHT": "<RAG_HYBRID_WEIGHT>",
        "RAG_WATCH": "<RAG_WATCH>",
        "HF_ENDPOINT": "<HF_ENDPOINT>",
        "HF_AUTO_MIRROR": "<HF_AUTO_MIRROR>",
        "HTTPS_PROXY": "<HTTPS_PROXY>",
        "HTTP_PROXY": "<HTTP_PROXY>"
      }
    }
  }
}

Replace the values in angle brackets with your own. Keys stay on your machine; we never see them.

Tools

This server runs locally, so its tools are listed by your MCP client once it starts.

Environment variables

Name	Description
`BASE_DIR` Optional	Base directory for document storage (defaults to current working directory). Ignored when BASE_DIRS is set.
`BASE_DIRS` Optional	JSON array of base directories (e.g. '["/a","/b"]'). Takes precedence over BASE_DIR.
`DB_PATH` Optional	Path to LanceDB database directory (defaults to ./lancedb/)
`CACHE_DIR` Optional	Directory where Transformers.js models are cached (defaults to ./models/)
`MODEL_NAME` Optional	Embedding model name (defaults to Xenova/all-MiniLM-L6-v2)
`MAX_FILE_SIZE` Optional	Maximum file size in bytes (defaults to 104857600 / 100MB)
`RAG_MAX_DISTANCE` Optional	Maximum distance threshold for filtering search results. Results with distance greater than this value will be excluded. Lower values mean stricter filtering (e.g., 0.5 for high relevance only)
`RAG_GROUPING` Optional	Grouping mode for quality filtering. 'similar' returns only the most similar group (stops at first distance jump). 'related' includes related groups (stops at second distance jump). Unset means no grouping filter
`RAG_MAX_FILES` Optional	Maximum number of files to keep in search results. Results are filtered to include only chunks from the top N best-scoring files. For example, 1 returns only the single best-matching file's chunks. Unset means no file filtering.
`CHUNK_MIN_LENGTH` Optional	Minimum chunk length in characters (1-10000, defaults to 50). Chunks shorter than this threshold are filtered out during ingestion.
`RAG_DEVICE` Optional	Execution device for the embedder (defaults to cpu). Passed straight to ONNX Runtime; see the Transformers.js device source for the supported backend names. If the requested device fails to initialize, the server throws an error.
`RAG_DTYPE` Optional	Embedding quantization dtype for the embedder (defaults to fp32). Opt-in and pass-through; accepts any dtype the chosen model provides (fp32, fp16, q8, int8, ...). If the model has no variant for the requested dtype, the server throws an error. Changing this changes the embedding space — re-ingest…
`RAG_HYBRID_WEIGHT` Optional	Keyword boost factor for hybrid search (0.0-1.0, defaults to 0.6). 0 means semantic similarity only; higher values increase the keyword-match contribution to the final score.
`RAG_WATCH` Optional	Enable automatic file watcher to detect changes and re-index modified files (true/1 to enable). Useful during active development.
`HF_ENDPOINT` Optional	Explicit HuggingFace endpoint URL to use for model downloads. When set, auto-mirror detection is skipped and this URL is used directly. Useful when you have a full-featured mirror.
`HF_AUTO_MIRROR` Optional	Enable automatic mirror detection for HuggingFace downloads (defaults to true). Set to "false" or "0" to disable and use huggingface.co directly. The mirror chain is: huggingface.co → hf-mirror.com → modelscope.cn.
`HTTPS_PROXY` Optional	HTTPS proxy URL for downloading embedding models (e.g. http://proxy:8080). Required only if behind a corporate proxy.
`HTTP_PROXY` Optional	HTTP proxy URL for downloading embedding models. Fallback if HTTPS_PROXY is not set.

Details

Package	`@damoqiongqiu/[email protected]` (NPM, stdio)
First published	23 Jul 2026
Last updated	23 Jul 2026
Registry name	`io.github.damoqiongqiu/mcp-local-rag`

More developer tools servers See all

Live status

Local install

Installed and run on your own machine, so there is no public endpoint to check.

Pair it with live search

Agents that use Mcp Local Rag often also need current facts. NestDaddy MCP gives your AI today's internet: web search, real-time breaking news and full page reading in one connector.

See NestDaddy MCP

For maintainers

[![NestDaddy MCP](https://nestdaddy.com/mcp/servers/io-github-damoqiongqiu-mcp-local-rag/badge.svg)](https://nestdaddy.com/mcp/servers/io-github-damoqiongqiu-mcp-local-rag)

Add the badge to your README. It turns green while your server passes our daily check. Listing details come from the official MCP registry; update them there.

Listing data from the official MCP registry. Status and tools from NestDaddy's daily check. Browse all MCP servers.