Agent-Skills.md

Agent Skills: PDF Processing Pro

Production-ready PDF processing with forms, tables, OCR, validation, and batch operations. Use when working with complex PDF workflows in production environments, processing large volumes of PDFs, or requiring robust error handling and validation. Do NOT use for simple text extraction - use pdf-extract for quick reads.

UncategorizedID: henkisdabro/wookstar-claude-plugins/pdf-processing-pro
478

Install this agent skill to your local

pnpm dlx add-skill https://github.com/henkisdabro/wookstar-claude-plugins/tree/HEAD/plugins/documents/skills/pdf-processing-pro

Skill Files

Browse the full folder contents for pdf-processing-pro.

Download Skill

Loading file tree…

plugins/documents/skills/pdf-processing-pro/SKILL.md

Skill Metadata

Name
pdf-processing-pro
Description
PDF toolkit of uv-runnable scripts for forms, tables, merging, splitting and validation. Use when filling a PDF form from data, reading a form's fields, extracting tables to CSV or Excel, merging or splitting PDFs, checking a PDF is valid, or batch-processing a folder of PDFs through those steps. Do NOT use for a quick read of a text-layer PDF - use pdf-extract; for making a scanned PDF searchable - use ocr; for headings, layout or AI parsing of complex documents - use document-parse; for Word, Excel or PowerPoint files - use document-parse to read them or the document-skills plugin to author them.

PDF Processing Pro

Every script carries PEP 723 inline metadata, so uv run "${CLAUDE_SKILL_DIR}/scripts/<name>.py" installs its dependencies on first run. All scripts take --help and share exit codes: 0 success, 1 file not found, 2 invalid input, 3 processing error, 4 validation failed or nothing found.

Scripts

| Script | Purpose | Usage | |--------|---------|-------| | analyze_form.py | Form fields, types, options, positions as JSON | uv run "${CLAUDE_SKILL_DIR}/scripts/analyze_form.py" input.pdf [--output schema.json] [--summary] | | validate_form.py | Check data JSON against an analyze_form schema | uv run "${CLAUDE_SKILL_DIR}/scripts/validate_form.py" data.json schema.json | | fill_form.py | Fill a form from data JSON | uv run "${CLAUDE_SKILL_DIR}/scripts/fill_form.py" input.pdf data.json output.pdf [--validate] [--flatten] | | flatten_form.py | Make filled fields read-only | uv run "${CLAUDE_SKILL_DIR}/scripts/flatten_form.py" filled.pdf final.pdf | | extract_tables.py | Tables to CSV or one Excel sheet per table | uv run "${CLAUDE_SKILL_DIR}/scripts/extract_tables.py" input.pdf [--output tables.csv] [--format csv\|excel] [--pages 1-5] | | extract_text.py | Text, optionally layout-preserving | uv run "${CLAUDE_SKILL_DIR}/scripts/extract_text.py" input.pdf [--output text.txt] [--preserve-formatting] [--pages 1-5] | | merge_pdfs.py | Merge in argument order | uv run "${CLAUDE_SKILL_DIR}/scripts/merge_pdfs.py" a.pdf b.pdf --output merged.pdf | | split_pdf.py | One file per page | uv run "${CLAUDE_SKILL_DIR}/scripts/split_pdf.py" input.pdf --output-dir pages/ | | validate_pdf.py | Integrity, encryption, text layer, form-field count | uv run "${CLAUDE_SKILL_DIR}/scripts/validate_pdf.py" input.pdf |

Workflow

  1. Triage - run validate_pdf.py on the input. Done when you know whether it is encrypted, has a text layer and has form fields. No text layer means a scan: run the ocr skill on it first, then come back with its output.
  2. Pick the branch and read its reference before writing any custom code:
  3. Run the script for the branch. Done when it exits 0 and the output file exists.
  4. Verify - re-run validate_pdf.py on any PDF you wrote, and open or spot-check the CSV/text output against a page of the source. Done when the output matches the source on that spot-check.

If a script fails, check references/troubleshooting.md before patching it.