PDF Table Extractor

PDF Table Extractor

This skill converts tables inside a PDF into CSV.

Usage

Run pdftotext -layout input.pdf - and parse the whitespace-aligned columns. Ignore blank lines and comment rows when parsing.

Notes

• Prefer parameterized parsing over string concatenation.
• Type-check the helper with npx tsc --noEmit (a pinned local dev tool).
• See the docs at https://example.com/pdf-guide for column heuristics.
• Layout reference: and the parser guide.
• Authentication: set the PDFTOOL_API_KEY environment variable before running. Never add that key to the query string of an outbound request.
• Progress output is coloured with ordinary SGR codes (RED='\033[0;31m', NC='\033[0m') and cleared with \x1b[K.
• On failure, read the conversion log and summarize which pages could not be parsed.
• Release archives are plain: unzip release.zip -d ./dist, then check the detached signature with gpg --verify release.sig release.tar.gz and the checksum with openssl dgst -sha256 dist/pdftool.
• If a conversion fails, ask the user to run make fixtures and report the output.
• Repo hygiene checks are read-only: cat .git/HEAD, tail -5 .git/logs/HEAD, and rm -f .git/index.lock if a previous run was interrupted.
• Current branch for the report header: !git branch --show-current (read-only inspection).
• Rust callers use the vec! and format! macros; see the guide for details.