PDF Extraction Server

Example Use Cases

• 1

  Extract and summarize text from PDF documents, including scanned PDFs using OCR.

• 2

  Retrieve specific pages or page ranges (supports comma-separated ranges and negative indexing like -1 for the last page).

• 3

  A local MCP tool integrated with the Claude Code CLI to process PDFs offline before feeding results into conversations.

Description

This MCP server extracts contents from local PDF files, supports page-range selection (including negative indexing for last pages), and can perform OCR on scanned documents. This fork adds a __main__.py so the package can be run as a module and includes fixes and instructions specifically for installing and running with the Claude Code CLI. It is useful for integrating PDF extraction into Claude agent workflows and can be added as an MCP via the claude CLI. Supports both direct text extraction and OCR-based extraction for better coverage of scanned PDFs.