Skip to content

NameetP/pdfmux

Developer Productivity

Self‑healing PDF extractor that audits output, re‑extracts problematic pages and supports multiple backends for clean LLM‑ready data

Track releases GitHub Website
Python Latest v1.6.4 · 29d ago Security brief →

Features

  • Per‑page confidence scoring and automatic re‑extraction of failures
  • Rule‑based routing to five specialized extractors (PyMuPDF, OpenDataLoader, RapidOCR, Docling, Surya) plus BYOK LLM fallback
  • CLI and Python API for single files, batch directories, streaming NDJSON, watching folders and CI‑friendly strict mode
  • Zero‑config defaults with optional extras via pip extras (OCR, tables, schemas, profiling, etc.)

Recent releases

View all 13 releases →
No immediate action
v1.6.4 Breaking risk

audit command + upsell line

Open
Review required
v1.6.3 Bug fix
Auth

Confidence fix

Open
No immediate action
v1.6.2 Maintenance

Routine maintenance and dependency updates.

Open
No immediate action
v1.6.1 Breaking risk

Strict threshold + batch_extract

Open
No immediate action
v1.6.0 New feature

Extraction backends + RTL + caching

Open

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

About

Stars
66
Forks
8
Languages
Python JavaScript Shell
Downloads/week
7 ↑550%
NPM Maintainers
1 Single npm maintainer
Contributors
5
View on GitHub View on npm Homepage

Install & Platforms

Install via
pip

Similar tools

signaturepdf
bentopdf
PSPDFKit/nutrient-dws-mcp-server
olgasafonova/mediawiki-mcp-server
ndjordjevic/pinrag

Alternative to

LlamaParse

Beta — feedback welcome: [email protected]