FICTIONAL EXAMPLE: not a real person, not real corpus. Part of the examples/ worked corpus (persona: Sam Rivers). See examples/README.md.

Monthly Review: 2026-02

~60 minutes. Synthesis. Led by Skeptic + Therapist + Coach as the rotating lens.

Patterns this month

February's shape: project:build-demo-and-wishlist-page opened on the 2nd and moved in small, real steps (build compiling, store page draft, decision:ADR-001 locking the price on the 18th). The habit:playtest-fridays cadence held two of three weeks (missed the 20th to a broken build) and its one clean February finding, that the tutorial takes four minutes to teach a 20-second mechanic, sat unfixed for a week after everyone already knew it. That's the real pattern: the easy 80% (page copy, a price decision) gets done on schedule; the known hard 20% (the tutorial cut) gets pushed a week at a time. area:health lost the same trade repeatedly: the last week of the month alone cost both scheduled runs and two late nights, and the weekly review found zero health evidence against that.

Candidate principles

aspirational:fund-health-like-a-load-bearing-area was added 2026-03-01, surfaced directly from the 2026-W09 weekly review's starvation finding. Confidence: medium, since one bad week is a data point, not yet a pattern, and the entry says so honestly (it's partly borrowed ambition, not yet lived). Falsification criterion as written: four consecutive weeks holding the standard, including one that overlaps real deadline pressure. February does not clear that bar on its own; it's the week that motivated writing the bar down. No promotion this month: there isn't a single week of counter-evidence yet, let alone four.

Full-stack coherence

A deeper pass than the weekly six-check: the ladder from project:build-demo-and- wishlist-page through goal:launch-paid-game-v1 to vision and principle:ship-before-its-ready stayed intact all month, and decision:ADR-001 is now load-bearing under the goal (a real price is what makes "100 paid units" a real target instead of a wish). area:health was starved at least the final week and, on the journal record, likely more of the month than one week's review caught, worth checking against the whole month's journal at the next weekly, not just the one sampled week. Capacity is still fine: one active project under one active goal is not over-scoped. No orphaned or contradicted artifacts found.

Status recommendations

No status changes recommended. project:build-demo-and-wishlist-page stays active (real movement, on target_date). goal:launch-paid-game-v1 stays active (still short of its own contested-early trigger of 150 wishlists by 2026-06-30). habit:playtest-fridays stays active (two kept, one missed for a real reason, not the two-in-a-row pattern that would lapse it).

The deeper question (Therapist)

The vision names "I stopped treating exhaustion as proof of seriousness" as part of who Sam is becoming, but February's actual behavior treats a late night as the proof the work mattered, and a run as the item that was optional. Worth sitting with, not resolving here: is skipping the run really about the deadline, or is it that finishing exhausted still feels like the only believable way to have earned the finish?

Carry forward

Keep the health question open into March rather than forcing an answer from one month's data. Track habit:playtest-fridays and area:health evidence with the same attention the project gets: right now one is logged weekly and the other only shows up when a review goes looking for it.