wiki-to-pdf

Pull FedWiki timeline pages and render formal executive-brief PDFs - one brief per timeline plus a combined dossier whose .

wiki-to-pdf

Turns the federated timeline wikis into formal PDF documents. Source of truth is the
live *.timeline.ide.earth subdomains, read over public GET (no cookie needed).

What it produces (into out/, inside this skill dir)

• <sub>-brief.pdf - one executive brief per timeline (cover -> overview -> In-brief -> Key Points / Action Items -> the five questions in detail).
• timeline-dossier.pdf - combined dossier: cover -> Contents -> Unified Summary (every In-brief, all 12 milestones) -> Milestones in Detail.
• model.json - the pulled, parsed content model (re-render offline with --model).

The Unified Summary section is deliberate: it is the shared semantic core (each
milestone compressed to its five answers) from which the detailed sections - and any
community-specific retelling produced by the understanding-compression skill - expand.

Run

python wiki_pdf.py                      # all 12 timelines + dossier -> out/
python wiki_pdf.py --sub agm            # one timeline brief
python wiki_pdf.py --no-dossier         # briefs only
python wiki_pdf.py --dump out/model.json   # pull + cache the model
python wiki_pdf.py --model out/model.json  # render from cache (no network)

How it works

1. Pull - for each timeline: the video/primary page (title, date object, video URL,
   Overview, Key Points, Action Items) + the 5 question pages (who-cares,
   who-is-involved, what-changes, proof--realness, becoming-real).
2. Parse - FedWiki story[] items -> structured record. For each question page it
   separates the ## In brief summary from the verbatim answer body and drops the
   navigation chrome (the intro line and # See block).
3. Render - ReportLab (Platypus). A minimal markdown->flowables converter handles
   headings, bullets, numbered lists, bold/italic/code, and wiki links.

Config

The timeline family, chronological order, video-slug map, and question slugs are at the
top of wiki_pdf.py (TIMELINES, QUESTIONS). Add/replace a milestone there.

Dependencies

• Python 3 + reportlab (pip install reportlab). No LaTeX / GTK / wkhtmltopdf needed.
• weasyprint is NOT used (its GTK libs are broken on this box); ReportLab is the engine.

Notes / gotchas

• Dates are read from each video page's date story item and shown verbatim ("date object true").
• The combined dossier paginates one milestone per section; ~64 pp for the current 12.
• If a question page is missing/renamed, that answer renders as "(missing: ...)" rather
  than failing the whole run.
• Pairs with fedwiki-cohere-sync (which keeps the source pages coherent) and feeds
  understanding-compression (which re-expresses the Unified Summary per community).

upstream: ~/.claude/skills (local workbench; not yet promoted into agentprivacy-skills-v5 — that promotion is a First-Person call) · author: Mitchell Travers · (⚔️⊥⿻⊥🧙)😊

← The Workbench Skills · Welcome Visitors