MCP Local RAG

MCPOpen SourceMIT68.8

by shinpr • Uncategorized

A fully local retrieval-augmented generation (RAG) server providing semantic search with keyword boost for technical documents.

Example Use Cases

1
Private, offline semantic search over local technical documents without sending data to external APIs.
2
Enhanced search precision by combining semantic similarity with keyword boosting for exact term matches.
3
Easy integration with AI coding tools like Cursor, Codex, and Claude Code for document ingestion and querying.

Description

MCP Local RAG enables developers to perform private, offline semantic search over their local documents with enhanced keyword matching for exact technical terms. It chunks documents semantically, generates embeddings locally using Transformers.js, and stores vectors in a local LanceDB database, requiring no cloud or API keys. The server supports multiple file formats and integrates easily with AI coding tools like Cursor, Codex, and Claude Code, offering zero-friction setup and strong privacy guarantees.

Capabilities(6 total)

Tools (6)

query_documents

Search ingested documents. Your query words are matched exactly (keyword search). Your query meaning is matched semantically (vector search). Preserve specific terms from the user. Add context if the query is ambiguous. Results include score (0 = most relevant, higher = less relevant).

query:string*limit:number

ingest_file

Ingest a document file (PDF, DOCX, TXT, MD) into the vector database for semantic search. File path must be an absolute path. Supports re-ingestion to update existing documents.

filePath:string*

ingest_data

Ingest content as a string, not from a file. Use for: fetched web pages (format: html), copied text (format: text), or markdown strings (format: markdown). The source identifier enables re-ingestion to update existing content. For files on disk, use ingest_file instead.

content:string*metadata:object*

delete_file

Delete a previously ingested file or data from the vector database. Use filePath for files ingested via ingest_file, or source for data ingested via ingest_data. Either filePath or source must be provided.

filePath:stringsource:string

list_files

List all files in BASE_DIR (PDF, DOCX, TXT, MD) and show which are ingested into the vector database. Also lists any other ingested items (web pages, clipboard content, etc.) that are outside BASE_DIR.

Quick Actions

View on GitHub

Security

Scanned 1 month(s) ago

Risk Level

MODERATE

Can modify files, state, or send data externally

Trust Score

B65/100

12/17 checks passed

Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.

Quick Stats

Service TypeMCP

Pricing ModelFree

Capabilities6 Tools / 0 Prompts / 0 Resources

Ownershinpr

CategoryUncategorized

DependenciesStandalone

Set Your Username