pdfmux is an open-source (MIT) PDF extraction router with a built-in
MCP server. Differentiator from existing PDF MCPs (MinerU, etc.):
instead of a single extractor, it classifies each page and routes to
the best backend (PyMuPDF / Docling / OCR / optional LLM), then
verifies output with per-page confidence scoring to prevent silent
extraction failures in RAG pipelines.
