batch-to-pdf

Batch to PDF

Combine a set of images, in order, into a single PDF document. Commonly used to produce a step-by-step tutorial after running annotate over a series of screenshots.

Originals-clean rule

Never modifies the input images. Only writes the single PDF at --output.

Just run the script

bash "$CLAUDE_PLUGIN_ROOT/scripts/to-pdf.sh" \
  --output "$OUT_PDF" \
  FILE [FILE ...]

Or with a glob (expanded by the shell, alphabetical order):

bash "$CLAUDE_PLUGIN_ROOT/scripts/to-pdf.sh" \
  --output "$OUT_PDF" \
  --glob "/path/to/annotated/*.png"

Order matters

The PDF page order follows the argument order. If the user has filenames like step-01-..., step-02-..., alphabetical glob expansion gives the right order. If they have unordered names, list the files explicitly in the order they want.

Backend selection

The script prefers img2pdf (lossless, no re-encoding for JPEGs — much smaller PDFs) and falls back to magick if img2pdf isn't installed. Run the install-deps skill once to get img2pdf.

Common workflow

1. screenshot × N → series of PNGs in ~/Pictures/Screenshots/
2. annotate × N → annotated copies in ~/Pictures/Screenshots/annotated/
3. batch-to-pdf → ~/Documents/tutorial.pdf from the annotated/ folder

Encourage the user to put the PDF outside the screenshots tree (e.g. ~/Documents/) so it doesn't get swept up by future glob runs.