Data Extract

Pull structured data out of documents and into a clean .xlsx + an audit report. The counterpart to data-tidy: tidy cleans data that's already roughly tabular; extract locates and pulls data from document-shaped sources (forms, multi-table PDFs, scans). Both share one engine, so extracted data is normalised and reported the same way.

Self-sufficient & local engine. No other toolkit needed. All processing — including OCR — happens on your machine: no cloud OCR, no external APIs, no third-party uploads. Note the AI agent driving the skill does send whatever it reads into its context to your AI provider; "never leaves the machine" is not claimed. See ../../PRINCIPLES.md#data-handling--pii-policy.

Shared engine. The cleanup primitives live at the plugin-root scripts/ (dataclean.py, ingest.py); this skill's scripts/extract.py adds them to the path.

Which mode?