pdf-page-count

PDF Page Count

Return the PDF's numeric page count and, when a limit was supplied, an explicit pass/fail comparison. Do not edit or rebuild the document unless the user also requested that separate work.

Quick Start

1. Resolve the exact PDF path and confirm the file exists.

2. Run the bundled counter from this skill directory:

   python3 scripts/count_pdf_pages.py <path-to-pdf>
   

3. Treat exit code 0 plus a single integer on stdout as success. Exit code 2 means none of the available backends could determine the count; report the dependency/error message instead of guessing.

4. Compare the integer to the user's rule:

   • maximum N: pass when count <= N;
   • minimum N: pass when count >= N;
   • exact N: pass when count == N.

One successful counter result for the exact output file completes counting. Do not render pages or run additional backends unless the file changed or a concrete discrepancy calls that result into question.

Output Contract

Report the file, page count, limit if any, and result. When over or under the limit, include the exact difference. Example: 27 pages; maximum 25; fail by 2 pages. Keep any requested caveat or next action, but omit backend narration on success.

If another task changed the source document, count the rebuilt output PDF, not the stale pre-edit artifact. If no rebuilt PDF exists, state that the count is for the currently supplied file.

Gotchas

• Do not infer pages from filename, file size, source-document pagination, or a viewer thumbnail count.
• Encrypted or malformed PDFs may fail every backend. Surface the failure; do not report zero pages.
• The script tries pypdf, pdfinfo, then macOS mdls. If no backend works, install pypdf or provide pdfinfo rather than rewriting the script during the counting task.
• A request to check a limit does not authorize layout compression or content cuts. Offer or perform PDF editing only when asked.