Workflows
PAD — Page A Day
365 daily context pages a year.
A daily generated page that fuses personal usage data, project context, industry news, AI updates, and world context into one structured digest. Builds a yearly journal as a side effect.
- 5 context streams aggregated
- Same shape every day
- 365 pages a year, evergreen
Why
Context decays fast. PAD freezes one structured snapshot per day — personal, project, industry, AI, world — so a year later you can still reconstruct the room you were standing in.
How
- Aggregate 5 streams via NotebookLM + RSS + Claude Code usage
- Render to the same fixed shape every day
- Persist to disk + Mongo for retrieval
Proof
- Cadence
- 1 page / day · 365 / yr
- Streams aggregated
- 5
- Average render time
- <3 min
PAD — five streams, one page, every day
Aggregate · Render · Persist
Hover or tap a node to see details.
FAQ
- What happens if one of the five streams isn't available that day?
- PAD degrades gracefully — the render proceeds with whichever streams returned and notes the gap inline. Yesterday's slot backfills the missing block so the daily shape stays intact.
- How long does a full PAD run take end to end?
- Under 3 minutes from cron-fire to persisted page on a typical day. Heavier industry days (paper drops, model releases) can push to 5 minutes; the bottleneck is research, not render.
- Can I retrieve and search past PADs?
- Every page is upserted to MongoDB AIDB.DAILY keyed by date, plus a full-text Markdown copy on disk. Both are queryable, and the year-end build merges them into a single yearly journal.
In production
- 365 daily PADs / year
One structured page every day — five context streams aggregated into a fixed daily shape.
- Year-end journal
365 PADs compile into a yearly journal — frozen daily context becomes a personal + business record.
- <3 min render
Cron-fire to persisted page in under three minutes on a typical day; ~5 min on heavy industry days.