Submit

mcp-docs-reader

@AIMIZING

A lightweight MCP server that loads PDF files, extracts and chunks their text, builds a semantic vector index, and returns relevant passages to Claude or other AI agents for document-based question answering.
Overview

What is mcp-docs-reader?

The mcp-docs-reader is a lightweight MCP (Model Context Protocol) server designed to load PDF files, extract and chunk their text, build a semantic vector index, and return relevant passages for document-based question answering.

How to use mcp-docs-reader?

To use mcp-docs-reader, install Claude Desktop, download the mcp-docs-reader project, set up the UV environment, configure Claude Desktop with the necessary settings, and run Claude Desktop to interact with your PDF documents.

Key features of mcp-docs-reader?

  • Loads and processes PDF documents from a local folder.
  • Extracts text and splits it into semantic chunks.
  • Generates vector embeddings using SentenceTransformer.
  • Builds a FAISS-based vector index for semantic search.
  • Retrieves top-k relevant chunks based on user queries.
  • Constructs prompts with relevant passages and questions for Claude.

Use cases of mcp-docs-reader?

  1. Answering questions based on local PDF documents.
  2. Summarizing key points from academic papers.
  3. Assisting in research by providing relevant document excerpts.

FAQ from mcp-docs-reader?

  • Can mcp-docs-reader process any PDF file?

Yes, it can process any PDF file located in the specified local folder.

  • Is there a specific setup required for Claude Desktop?

Yes, you need to configure Claude Desktop to recognize the mcp-docs-reader settings.

  • What is the purpose of the semantic vector index?

The semantic vector index allows for efficient retrieval of relevant text passages based on user queries.

© 2025 MCP.so. All rights reserved.

Build with ShipAny.