This is a solid general-purpose PDF extraction skill that handles the usual challenges: text-based versus scanned documents, table reconstruction, form field extraction, and multi-column layouts. It'll preserve document structure, flag unreliable extractions, and output to Markdown, JSON, or CSV depending on what you need. The skill includes specific patterns for common document types like invoices, legal contracts, and financial reports, which saves you from reinventing extraction logic each time. It's honest about OCR limitations and won't silently mess up table alignments. If you regularly wrestle with PDF parsing instead of doing actual work with the data inside, this handles the boring structural stuff so you can focus on analysis.
npx skills add https://github.com/rightnow-ai/openfang --skill pdf-reader