mcp-docs-rag
If you are the rightful owner of mcp-docs-rag and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
This is a TypeScript-based MCP server implementing a RAG system for documents stored in a local directory.
The MCP Server for RAG (Retrieval-Augmented Generation) is designed to facilitate document querying using LLMs with context from locally stored repositories and text files. It supports documents in the form of Git repositories or plain text files, allowing users to access and query these documents efficiently. The server is built using TypeScript and leverages Google's Gemini API for document indexing and querying. Users can add documents via Git repositories or text files, and query them using RAG to receive AI-generated responses with contextual information. The server is configurable and can be integrated with platforms like Claude Desktop for enhanced document management and retrieval.
Features
- Resources: List and access documents via `docs://` URIs, supporting Git repositories and text files with plain text mime type.
- Tools: Includes tools like `list_documents`, `rag_query`, `add_git_repository`, and `add_text_file` for document management and querying.
- Prompts: Provides `guide_documents_usage` for guiding users on document and RAG functionality usage.
Tools
list_documents
Lists all available documents in the DOCS_PATH directory, showing a formatted list and total count.
rag_query
Queries documents using RAG, taking document_id and query as parameters to return AI-generated responses.
add_git_repository
Clones a Git repository to the docs directory with optional sparse checkout, taking repository_url and optional parameters.
add_text_file
Downloads a text file to the docs directory using wget, taking file_url as a parameter.
