PDF documents

Use this Skill for PDF extraction, generation, page operations, forms, metadata, attachments, encryption, and visual review.

Start here

Call skill_view for this Skill first and use the returned baseDirectory for all bundled helpers under <baseDirectory>/scripts.

Default to a new output file. Do not overwrite or remove source PDFs unless the user explicitly asks. Inspect an input before transforming it, and preserve page order, orientation, dimensions, annotations, and form intent unless the request says otherwise.

Common dependencies:

python3 -m pip install pypdf reportlab pdfplumber

Page rendering uses pypdfium2 when available and can fall back to Poppler's pdftoppm. Check dependencies first; do not install them silently.

Helpers

• pdf_read.py: extract text and document information.
• pdf_page_image.py: render selected pages to images.
• pdf_create.py: create a PDF from structured input.
• pdf_merge.py and pdf_split.py: page composition.
• pdf_fill_form.py, pdf_form_layout.py, and pdf_make_form.py: AcroForm inspection, filling, and creation.
• pdf_watermark.py and pdf_stamp.py: page overlays.
• pdf_meta.py: metadata and attachment operations.
• pdf_secure.py: encrypt or decrypt using password files.

Run helpers with terminal_exec, using python3 as the command and an argument array. Use --help to inspect exact arguments.

python3 <baseDirectory>/scripts/pdf_read.py input.pdf --json
python3 <baseDirectory>/scripts/pdf_page_image.py input.pdf -o rendered --pages 1-3
python3 <baseDirectory>/scripts/pdf_merge.py first.pdf second.pdf -o merged.pdf
python3 <baseDirectory>/scripts/pdf_split.py input.pdf -o pages

For form work, read references/forms.md through skill_view before editing.

Workflow