Guides

This section stores knowledge pages generated from source PDFs and then reviewed by an editor.

Operating flow

1. Drop a source PDF into pdf-to-images/inbox/.
2. Run the intake processor.
3. Review the generated page under this section.
4. Edit the summary and keep or trim the extracted page images as needed.

What the automation creates

For each incoming PDF, the workflow creates a numbered folder in this section containing:

• an index.md knowledge page ready for editorial review;
• an images/ folder with one rendered image per PDF page and page-local searchable text in the generated document;
• an extracted-text.txt file with the page text pulled from the PDF for audit or downstream reuse;
• a manifest.json file with basic processing metadata.

Editorial note

The generated page is intended as a starting point, not a final publication artifact. The script handles ingestion, traceability, and layout scaffolding; a human should still tighten the narrative before treating the page as authoritative knowledge.