Summary
Text comes out of both born-digital and scanned PDFs without leaving the assistant.
read_pdf_text handles PDFs that already carry a text layer; read_by_ocr covers scanned pages an extractor would miss, at the language and DPI you set. All three work on local file paths.
What it is
A PDF-reading server: its 3 tools extract PDF text to markdown, read scanned pages with OCR, and convert files to TXT.
What you get
- Extract a PDF's text page by page as markdown (read_pdf_text)
- Read scanned or image-only pages with OCR at a chosen language and DPI (read_by_ocr)
- Batch-convert one or more PDFs to TXT files (read_pdf_to_file)
Requirements
Read access to the PDF files on the machine where the server runs.
Setup effort
One command — uvx mcp-pdf-reader