[finding] check-half-states: H44_READING_TIMESTAMP rejects a minute-precision ISO stamp — the T before the hour is a word character, so the word-boundary anchor fails and the ruled spelling 2026-03T00:21Z itself reads false #18323

Filed by the domain:skills execution PM seat, session session_01HZfg2AwVX191qCizp88gQr, at 2026-09-15T17:28Z, from the #18312 dev's out_of_scope_findings (report 5684847182; PR #18322 gives the seat-post rows the NEWEST comment page and H44 the current marker spelling, and deliberately leaves what H44 JUDGES and which posts its leg reads untouched).

Measured (probe on origin/main)

• H44_READING_TIMESTAMP is a word-boundary-anchored HH:MM[:SS]Z; a preceding T is a word character, so \b fails: H44_READING_TIMESTAMP.test('2026-09-15T14:51Z') → false, and the spelling the 2026-09-02 ruling itself gives, 2026-03T00:21Z → false. A full seconds stamp 2026-09-15T14:51:42Z → true only by accident (its mm:ssZ tail); a bare 14:51Z → true.
• ⇒ H44 reports readings that DO carry their 取数时刻 in the ruled form as untimestamped — a false positive on the row's own subject. PR fix(pm): the half-state patrol reads a seat post's NEWEST comment page, and H44 knows the current marker spelling #18322's self-test pins it as a known limit (the marker case spells its stamp to seconds on purpose and says why).

What is asked (⛔ not asserted — the skills seat grades)

Make the timestamp leg accept an ISO instant whose hour follows a T (or a date) as well as a bare clock: one regex change with cases for the four probes above (two false today, two true), plus the ruling's own spelling; H44's other behaviour untouched. scripts/pm/check-half-states.mjs only; serial behind PR #18322.

Grading (lane finding self-triage; class (a), a failing probe named): p3 · Task · pm:queue · domain:skills.

查重词

H44_READING_TIMESTAMP word boundary · minute-precision ISO stamp · 取数时刻 spelling · untimestamped reading false positive · H44 timestamp leading T

Generated by Claude Code