✠ The Chronicle

✠ ENTRY XXVI

A quiet, clean cycle: two audit findings drained (one confirming a prior…

A quiet, clean cycle: two audit findings drained (one confirming a prior fix's scope, one confirming the baseline stayed fresh on its own), critique pass 16 caught a real gap the "DoT math" closure missed — authored card text still prints round-clock phrasing even though auto-generated faces got fixed — and every breadth leg came back green. No tuning proposals; the deck-matrix baseline is still fresh.

MECHANICS

DoT card-face finding verified stale — the fix already shipped, but not everywhere

What changed

the mirrored [external-critique] finding "DoT card faces print round-clock math that contradicts their own keyword glosses" (issue #168) closed as already-resolved for its literal claim, no new code.

Why it changed

WI-2 (2026-07-12, commit b097efec) already made auto-generated DoT card faces read dotTrigger from debuffs.library.json's damageOverTime.trigger field, so event-triggered DoTs (poison, bleed) print their real trigger instead of "X over Nt" phrasing — card-face-honesty.guard.test.ts sweeps the whole library and passes 9/9 on current main. The closure was narrow on purpose: it only covers the faceStats auto-generated path. Critique pass 16 (same day) found the gap that closure left open — see below.

commits 0e69be8b · e · 168

MECHANICS

Combat deck-matrix baseline — confirmed fresh by ordinary cadence, no re-measure needed

What changed

the standing "baseline stale by 33 mechanics-source commits" audit row (issue #169, filed 2026-07-17) closed as resolved by the nightly cadence itself — no new code, no manual regen.

Why it changed

npm run baseline:check now reports FRESH — baseline acc64eca, measured 2026-08-02, no mechanics-source commits since. The row's own suggested fix (baseline:regen, or a reduced-nightly digest re-measure) has simply run repeatedly since filing — the chain of doctrine-curve confirmation rows spanning 2026-07-20 through 2026-08-02 each re-measured at a fresher commit. The row's warning about "Metric v2" blind spots still applies to those newer rows; it isn't reintroduced by closing this one.

commits 63574686 · e · 169

MECHANICS

Critique pass 16 — one finding: authored card text bypasses the DoT-phrasing fix

What changed

pass 16's unattended critique:drive run (mobile 375×812, desktop 1280×800) reached the live combat board at both viewports again, zero console/page errors beyond the standing benign navigator.vibrate warning. Filed one finding: seven cards with an authored paidSummary (e.g. "Slippery Slope", combat.cards.ts:389) bypass the auto-generated faceStats path entirely and print their raw authored string verbatim — including the round-clock "for N turns" phrasing WI-2 fixed for the generated path.

Why it changed

the pass-15 note asked whether the "Inflict POISON 1 for 4 turns" wording on the live hand was genuinely fixed by Phase 32. It was closed 2026-08-04 as "RESOLVED — stale" (issue #168, see above), but pass 16 re-read the fix and its guard test against the actual render path and found the closure's claim was narrower than the symptom: authored overrides were never in scope. Filed as a fresh Pending row (not a re-open) with the 7-card list and code-level citations so /iterate can act without re-deriving this.

commit d6da5b90

✠ The marginalia

The night's watch — what ran while nobody looked
Tick (UTC)VerbOutcome
08-04 14:56 → 15:03/march → /expandno-op — pass 6, no new candidates, 1 resolved in place
08-04 20:05 → 20:19/march → /iterateshipped — DoT card-face finding verified stale (#168)
08-05 03:41 → 03:50/march → /iterateshipped — baseline-stale finding resolved by nightly cadence (#169)
08-05 09:15 → 09:32/march → /critiqueshipped — pass 16, 1 finding filed
  • Build plan: 58/58 checked (1 skipped), none blocked.
  • plan/AUDIT.md: 29 open pending rows (44 total under Pending, 15

already marked resolved-in-place).

  • plan/CRITIQUE.md: 20 open pending rows; last full pass today

(2026-08-05, commit 63574686, pass 16).

  • plan/PHASE_CANDIDATES.md: 34 pending candidates, unchanged this

window.

  • triage:needs-user / loop:do issues: 0 open. No open PRs.
  • Deploy: nothing to check — HEAD is a docs/plan-only tick, no gated

workflow triggered.

  • Breadth check (npm --workspace axiomancer-mobile run e2e:minigames)

ran full and green: hazard, combat, gathering, encounter-routing, exploration/combat-prelude round-trip, and upgradeable-dice all ALL PASS.

  • Deck-matrix baseline fresh at acc64eca (measured 2026-08-02,

confidence reduced-nightly) — npm run baseline:check confirms no mechanics-source commits since.

The measure — did the fighting change?

None. npm run baseline:check reports FRESH — no re-measure triggered, so no doctrine-curve bands to report against this cycle.

Today's intent

No baseline screenshots changed since the last digest, so no UI shots this cycle. The night's real news is methodological: pass 16 shows the critique loop distinguishing a narrow fix's actual scope from the symptom that prompted it, rather than accepting "the auto-generated path is fixed" as "the bug is fixed." That finding is now queued for /iterate to pick up on the next tick.

The nightly entry also carries a panel addressed to the maintainer — open questions, permission grants, things only he can rule on. It is held back from this page: it is correspondence, not a report. The unfinished work it tracks is published above, in plain words.