agentsclimarketplace

Pdf image converter

Skill bybaochaule/pdf-image-converter

This Agent Skill converts PDF pages into PNG, JPEG, or JPG image files. It is designed for page rendering, preview generation, and batch PDF-to-image conversion.

Install
npx -y skills add bybaochaule/pdf-image-converter

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use this skill to convert PDF pages into PNG or JPEG image files. Trigger when the user asks to render PDFs, export PDF pages as images, convert PDF to PNG, JPEG, or JPG, create page previews, or batch-convert PDF files. Do not use it for OCR, text extraction, PDF editing, or image-to-PDF conversion unless explicitly requested.

SKILL.md

4.9 KB, ~1.0k tokens by cl100k_base, as published. Nobody here has run it

PDF Image Converter

Purpose

Convert one or more PDF files into raster image files, with one output image per selected page. Preserve the visible page appearance as closely as possible and make the conversion reproducible with explicit format, DPI, page range, naming, and verification steps.

When to use

Use this skill when the user asks to:

  • Convert a PDF, multiple PDFs, or selected PDF pages to PNG, JPEG, or JPG images.
  • Render PDF pages as page previews, thumbnails, screenshots, or web-ready images.
  • Batch-convert uploaded PDF files into image folders or a ZIP archive.
  • Control image output settings such as DPI, page range, JPEG quality, or output directory.

Do not use

Do not use this skill when:

  • The user needs OCR, text extraction, table extraction, form extraction, or semantic document understanding.
  • The user wants to edit, merge, split, redact, repair, compress, or otherwise modify a PDF.
  • The user wants to convert images or office documents into a PDF.
  • The input is not a PDF file and no PDF conversion step was requested.
  • The file appears encrypted, restricted, or private and the user has not provided authorization or a valid password.

Required inputs

Use information already provided by the user when possible. Required or defaulted inputs are:

  • Source PDF file path or uploaded PDF file.
  • Target image format: png, jpeg, or jpg; default to png when unspecified.
  • Page selection: all pages by default, or a 1-indexed page list such as 1,3-5.
  • DPI or resolution: default to 200 DPI for general use; use lower DPI for very large PDFs and higher DPI for print-quality output.
  • Output location: default to <pdf-stem>_images/ and one image per page.
  • JPEG quality when using JPEG/JPG: default to 90.
  • Password only when the PDF is encrypted and the user is authorized to open it.

Workflow

  1. Confirm the request is a PDF-to-image conversion task and identify the source PDF files.

  2. Resolve missing settings safely:

    • Use png unless the user asked for jpeg or jpg.
    • Use 200 DPI unless the user requested another resolution.
    • Convert all pages unless the user requested a subset.
    • Name outputs as <pdf-stem>_page_0001.<ext>.

  3. Create a clean output folder. For multiple PDFs, place each PDF's images in a separate subfolder.

  4. Run the deterministic helper script when local execution is available:

    python scripts/pdf_to_images.py input.pdf --format png --dpi 200 --out-dir output_images

    For JPEG:

    python scripts/pdf_to_images.py input.pdf --format jpeg --quality 90 --dpi 200 --out-dir output_images

    For selected pages:

    python scripts/pdf_to_images.py input.pdf --pages 1,3-5 --format png --dpi 200 --out-dir output_images

  5. Prefer the helper script's PyMuPDF renderer. If PyMuPDF is unavailable, the helper falls back to Poppler pdftoppm when installed.

  6. Verify the output:

    • Count the generated image files and compare them with the selected page count.
    • Open or inspect at least the first generated image.
    • Confirm naming, extension, DPI choice, and output directory.

  7. Package images as a ZIP only when requested or when many output files would be inconvenient to return individually.

  8. Return links to the generated image files or ZIP archive, plus a short summary of format, DPI, pages converted, and any warnings.

Output format

Return a concise final answer with:

  • A link to the output folder, image files, or ZIP archive.
  • Format, DPI, page range, and file count.
  • Any assumptions, skipped files, encryption/password issues, or conversion warnings.

Quality checklist

Before finalizing:

  • The input file paths exist and are PDF files.
  • The requested format is normalized to PNG or JPEG/JPG.
  • The output directory exists and contains the expected number of image files.
  • Each image file has a stable, page-numbered name.
  • The first output image opens successfully.
  • Large PDFs were handled with an appropriate DPI or page subset.
  • Temporary files were not returned as final artifacts.
  • Validation errors are reported clearly instead of being hidden.

Safety and privacy

  • Do not expose secrets, credentials, private document contents, or passwords in final responses.
  • Do not bypass PDF access restrictions or remove encryption.
  • Do not run destructive shell commands or delete user files.
  • Avoid network access; conversion should be local and deterministic.
  • Do not collect passwords unless the PDF is encrypted and the user confirms authorization.
  • Keep generated files in the requested output folder or a safe working directory.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.