MCP Local RAG
mcp-local-rag
Local search over your documents, semantic and keyword, with no keys, Docker or external database. Works offline after the first model download
- 414 ★ on the repository
- Last commit
- MIT
Minimal access
Why this level
- The index is on disk and texts do not go out
- The agent sees all documents in BASE_DIR
Install
- Works in
Manual install
claude mcp add local-rag --scope user --env BASE_DIR=/absolute/path/to/your/documents -- npx -y mcp-local-ragIndexing: npx mcp-local-rag ingest ./docs/. Query: npx mcp-local-rag query "text".
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- The index is on disk and texts do not go out
- The agent sees all documents in BASE_DIR
This is third-party code. Review the repository files before installing.
What it does
MCP Local RAG searches your documents both by meaning and by keywords, locally and with no API keys, Docker or external database. It works offline after the first model download. Connection: `claude mcp add local-rag --scope user --env BASE_DIR=/absolute/path/to/your/documents -- npx -y mcp-local-rag`. The commands `npx mcp-local-rag ingest ./docs/`, `query` and `sync` work from the terminal. Version 0.21.0 from 03.10.2026, MIT license.
Who it is for. For people who want to search their documents from an agent without a cloud and keys.
Good fit when
- You need local document search with no API keys
- You need meaning and keyword search
- You need offline work after the first model download
Not a fit when
- You need cloud search with shared access
- You have no room for the model: it is downloaded once
Example request
Find mentions of API authorization in my documents and list the files with quotesMCP
- Transport
- stdio
- Authentication
- not required
Environment variables
BASE_DIRrequired- Environment variable BASE_DIR, the value is set in the client or a .env file
Limitations
The document index is stored locally and texts do not go out, so access is minimal. The model is downloaded once at the first launch. Choose the BASE_DIR folder knowingly: the agent will see all its documents.
How to disable. Remove the server entry from the client and the index in the program data folder.
FAQ
What does MCP Local RAG do?
It searches your documents by meaning and keywords locally.
Are keys needed?
No, neither API keys nor Docker nor an external database.
Does it work offline?
Yes, after the first model download.
Discussion
No questions or reviews yet.
Related
MCP serversOn-premises RAG in containers with an MCP server: the agent searches your documents while data stays local
haiku.rag
haiku-rag
Agentic RAG over your documents with citations (page, heading) and a built-in LanceDB. The MCP server starts with haiku-rag mcp --stdio
A CLI engine that compiles documents into a local structured wiki: a transparent alternative to plain RAG, reachable by an agent over MCP
A CLI and MCP server that extracts structured knowledge from documents: graphs, hypergraphs, and fact search with sources
Official MongoDB MCP server: queries and aggregations, index management, Atlas clusters and local deployments
An MCP server for searching and downloading academic papers from arXiv, PubMed, bioRxiv, Semantic Scholar and other open sources
README in short
The README describes local document search, semantic and keyword, with no keys and no external database, connection through npx. MIT license.