Practical OCR, with receipts.

Extract text from receipt photos in Python with three lines of code — handles crumple, blur and skew. Includes parsing totals and a full runnable example.

Turn any PDF — scanned or born-digital — into clean markdown in Python: headings, reading order and HTML tables preserved. Perfect for RAG and LLM pipelines.

Read PDF and scanned invoices in Node.js with a few lines: full text or markdown with line-item tables, then parse amounts. Runnable example included.

Turn a scanned PDF into text or markdown from your terminal with a single curl command. Plus jq recipes for extracting pages and piping into other tools.