docs(skills): brainstorm-with-docs — перечитывание реагирует по уровню находки

Текстовые огрехи горячее ревью правит молча по месту; находку уровня
решений (противоречие, дыра в охвате) предъявляет владельцу, не исправляя
и не замалчивая. Довод владельца: текстовые находки видны и неопасны, а
противоречия при перечитывании не раз оказывались важными — плоский запрет
переоткрывать решения заставлял бы молчать или чинить мимо владельца.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
This commit is contained in:
Dmitry DementievandClaude Opus 5 committed 2026-08-01 19:12:59 +03:00
1 parent a95dc7988e
commit 9ce4c6f769
2 files changed
+17 -7

No files matched your search

@@ -156,9 +156,16 @@ shaped by three inputs:
Landed shape — two tiers: `<self_review>` runs at step 6 on the plan (can send Landed shape — two tiers: `<self_review>` runs at step 6 on the plan (can send
the run back to the interview; no file exists yet), step 7 writes the file once the run back to the interview; no file exists yet), step 7 writes the file once
directly, then re-reads it end-to-end as a reader and fixes text-level defects directly, then re-reads it end-to-end as a reader. Unlike upstream, the review
in place (does not reopen decisions). Unlike upstream, the review itself is itself is kept — `to-spec` has none; ours has content worth keeping.
kept — `to-spec` has none; ours has content worth keeping.
Follow-up (same day): the re-read's reaction now splits by the level of the
find. Text-level — fix in place, silently. Decision-level (contradiction with
docs or between sections, coverage hole, unhandled scenario) — present to the
user, neither silently fixed nor suppressed. Owner's evidence: text-level finds
are visible and harmless, but contradictions caught on re-read have mattered
more than once; a flat "don't reopen decisions" ban would force the agent to
either suppress such a find or fix it past the owner.
## v2 — remaining ## v2 — remaining
@@ -244,10 +244,13 @@ Otherwise use the conventional homes — `CONTEXT.md` at the repo root, ADRs in
as part of step 6 rather than guessing. Create directories lazily. as part of step 6 rather than guessing. Create directories lazily.
After writing, re-read the written file end-to-end as its reader, not its After writing, re-read the written file end-to-end as its reader, not its
author, and fix what you find in place — wording, structure drift, sections author. React by the level of what you find. Text-level defects — wording,
that disagree with each other, template mismatches. This pass is cheap and structure drift, template mismatches — fix in place, silently. A
regularly catches real defects; it polishes text, it does not reopen decisions decision-level find — a contradiction with existing docs or between the
(those passed `<self_review>` at step 6). artifact's own sections, a hole in coverage, an unhandled scenario — present
to the user instead: neither fix it silently nor suppress it. Reopening a
decision is the user's call. This re-read is cheap and regularly catches both
kinds.
Then go to `<stop_condition>`. Then go to `<stop_condition>`.