| name | dark-pdf-studio |
| description | Convert PDFs, document files, or images into dark-background reading PDFs with a compact input-to-output workflow. Use when Codex needs to dark-theme a PDF, normalize docs or images into a dark PDF, build or refine a tiny dark-mode export utility, or troubleshoot a dark PDF conversion flow with downloadable output. |
Dark PDF Studio
Use this skill when the user wants a simple dark-PDF workflow: one input file, one output file, dark background, downloadable result.
Quick Start
- Prefer PDF input for layout fidelity.
- Run
bash scripts/run_dark_pdf_studio.sh doctor.
- Run
bash scripts/run_dark_pdf_studio.sh inspect.
- Use
bash scripts/run_dark_pdf_studio.sh convert --input input.pdf --output output-dark.pdf.
- Use
--theme graphite, --theme midnight, or --theme navy for the background tone.
- For document inputs such as
.docx or .rtf, let the script normalize them through textutil first on macOS, or LibreOffice when needed.
- Use
bash scripts/run_dark_pdf_studio.sh validate after edits to confirm the converter still compiles and exposes its CLI.
- If the user wants a small app, keep the UI compact:
- choose or drop one file
- pick one dark theme
- export
- open or reveal output only after export exists
Workflow
Input Rules
- Best input: PDF
- Also supported:
.doc
.docx
.rtf
.odt
.txt
.md
.html
.htm
- common images including
.png, .jpg, .webp, .gif, .heic
- Prefer converting non-PDF documents into PDF before dark processing so the export contract stays consistent.
Output Rules
- Default output is a dark-background PDF.
- Keep the surface simple:
- one input
- one output
- one theme choice
- Prefer an essentials-only utility surface:
- compact header
- source picker / drop target
- theme picker
- export button
- open / reveal only when there is output
- Treat “downloadable” as a local exported file path that can be opened or shared immediately.
Editing Guidance
- Use
scripts/run_dark_pdf_studio.sh for the normal doctor, inspect, convert, and validate flow.
- Use
scripts/dark_pdf.py for the actual conversion path when you need the lower-level CLI directly.
- Keep the conversion deterministic and easy to explain.
- If a document backend is missing, fail with a clear dependency message instead of pretending the conversion succeeded.
- Prefer
textutil on macOS for document-to-text normalization before falling back to LibreOffice.
- Prefer readable high-contrast results over preserving every original page color exactly.
Validation
- Run
bash scripts/run_dark_pdf_studio.sh validate after edits.
- If you need to prove end-to-end conversion, smoke-test one real PDF through the runner and confirm the output file exists.
- Run
python3 scripts/dark_pdf.py --help only as a fallback when debugging the wrapper itself.
- If the user provides a real file, smoke-test one conversion and confirm the output path exists.
Resources
scripts/run_dark_pdf_studio.sh: repo-native doctor, inspect, convert, and validate helper for the dark PDF workflow.
scripts/dark_pdf.py: CLI converter for PDF, doc, and image inputs into dark-background PDF output.
references/workflow-map.md: input contract, backend expectations, and output behavior.