io.github.penny4nonsense/mcp-pymupdf

Extract text from a PDF file or URL. Reads up to max_pages pages and max_bytes bytes of content. Scanned or image-based PDFs will be flagged as unreadable. Args: source: Local file path or URL to the PDF. max_pages: Maximum number of pages to read. Defaults to 50. max_bytes: Maximum output size in bytes. Defaults to 200000. Returns: Extracted text with page markers, or an error message.

Extract text from a specific page range of a PDF. Pages are 1-indexed. If end_page is not specified, only start_page is read. Args: source: Local file path or URL to the PDF. start_page: First page to read (1-indexed). Defaults to 1. end_page: Last page to read (1-indexed, inclusive). Defaults to start_page if not specified. max_bytes: Maximum output size in bytes. Defaults to 200000. Returns: Extracted text from the specified pages.

Retrieve metadata from a PDF without reading its full text. Returns title, author, subject, keywords, creator, producer, creation date, and page count. Args: source: Local file path or URL to the PDF. Returns: Formatted metadata string.