# VERTO public website capture

Start with sitemap.html. It works offline and its live-page links remain distinct from the local Markdown links.

Captured public source only, on 29 October 2026 (Australia/Sydney). See IA-report.md and summary.json for exact scope/counts.

The archive includes raw public HTML, extracted main-content Markdown, navigation, the published sitemap, metadata and link relationships. It is not a full media/document mirror or a CMS export.

Build order (Python environment requires advertools, requests, lxml, beautifulsoup4 and markdownify):

1. scripts/crawl.py, with scripts/ on PYTHONPATH and the environment bin/ on PATH. Do not rerun into an existing crawl output: advertools appends.
2. scripts/process_capture.py
3. scripts/build_sitemap.py

Navigation was recovered independently and is preserved under navigation/. No website writes or form submissions were made.

Raw source sitemap locs are kept unchanged; retrieval-only entity decoding is explicitly labelled. A robots-disallowed PDF is inventoried only if linked and is not fetched.

See verification.json for coverage reconciliation, archive integrity and browser checks.