You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Standards live in per-concern files loaded by progressive disclosure; AGENTS.md holds no review
rules; REVIEW.md and other tool-specific files point at the per-concern files or are generated from
them, opt-in.
Decisions
No single coding-standards file and no review rules inside AGENTS.md; standards split per concern,
loaded by progressive disclosure; tool-specific files point at that source or are generated from it
(D97, Q150, Q570). Origin: ours (rejects Matt's single file and Lauren's rule table in the
instruction file).
REVIEW.md: generation is opt-in, keeps a hand-maintained section; the nit cap lives in its
per-concern file (Q570). The Q565 link-following probe is withdrawn (GUIDE 122): this repository's
reviewers are the GitHub Actions review lanes running the review plugin, which follow REVIEW.md
citations, so no probe gates this thread.
Gardening (retro, post-merge sampling, cleanup, scheduled) keeps the standards current (D97).
REVIEW.md syncs from melodic-software/standards, so the change there is another-repository work
(owner stop).
Key interfaces
Variation verdicts: "REVIEW.md and other tool-specific review files: contain now; point at, or opt-in
generated from, per-concern files (Q570)."
Slices, filed when this phase starts: per-concern-split, review-md.
Done when
Every per-concern standards file exists and REVIEW.md links each one (the folder does not exist
on main).
The drift check exits non-zero after a hand edit inside a generated block of a fixture copy and
0 after regeneration.
Every slice sub-issue is closed by a merged PR.
Out of scope
Loading the standards index by default (parked item).
Proposed work class: work-class: structural
Basis: automated intake triage; the lanes App applied this class.
Size: Large
Complexity: Medium
Uncertainty: Known
Flags: none
Blocked by: none
Decision recorded. This issue's - [ ] criteria are already in its body, written by a listed human. The owner amended D97/Q150 on 2026-10-16 to allow one pointer line in AGENTS.md. Nothing new is decided here. [...]
[quote truncated; the full decision is at the link]
The decision settles the open questions: standards go into per-concern files, AGENTS.md keeps one pointer line, and the issue closes through its sub-issues (per-concern-split, review-md). Nothing is left to ask.
I found no open duplicates. The dependency #7571 isn't in the open-issue list I was given, so I couldn't confirm its state. Part of the work is in melodic-software/standards, which is a separate repository.
D97/Q150 (GUIDE 97) is amended to allow one pointer line in CLAUDE.md/AGENTS.md to the per-concern review files. Reason: local /code-review reads CLAUDE.md, and its docs say of REVIEW.md: "It doesn't read REVIEW.md" (https://code.claude.com/docs/en/code-review-local). Without a pointer, a local review sees none of this repository's review rules. The line is a pointer, not a rule, so Q150's "no review rules inside AGENTS.md" holds.
The rule list now under AGENTS.md ## Code Review Rules moves into the per-concern files, leaving only the pointer line.
The gap is narrow: under Q500 (GUIDE 75) a repository with a CI lane runs no local review, and the CI lanes follow REVIEW.md citations (GUIDE 122). It affects ad hoc local sessions and consumers set to review_venue: local.
A suggestion from the ADR review consensus, not an owner decision: load the per-concern files by path scope, so each enters context only when a file it covers is touched.
This thread already puts the nit cap in its per-concern file (Q570). The article's REVIEW.md traits it lacks (04 C18-C20), which belong in those files:
Keep it short.
Say what failure looks like in this repository (for docs "it lies", for config "it misroutes").
List paths to skip.
Ask for a file and line as evidence instead of a guess from a name.
Ask for a summary that starts with a count and says plainly when there is nothing to fix.
The article also checks separately whether the reviewer follows REVIEW.md (04 C17).
Background only: managed Code Review is not used here (GUIDE 122, Q594). Q570's generated REVIEW.md stays an opt-in for consumers on that service.
Against the originating sources. This thread rejects Matt's single standards file and Lauren's rule table in the instruction file. The article agrees with pointing out of a short CLAUDE.md (04 C14); its reviewer also flags code that breaks a CLAUDE.md convention (04 C15). The pointer line meets 04 C15 without putting rules in the instruction file.
Released for re-triage under the owner's 2026-10-24 rule: an issue stops for a person only for an intent or vision question, a security call, a host or attended run, or spend outside prepaid credit. The hold label came off and intake triage runs again now; it re-applies a hold, with a numbered question, when one of those applies.
To hold this issue again, add needs-human (or status: needs-decision) and say why in a comment.
Automated split by the lanes App. Admission waits on this issue while a sub-issue is open; work-unblock closes it as completed once every sub-issue is closed by a merge or a listed human.
Issue Triage session: removing agent-ready from this split parent; it waits on its open sub-issues #10231 and #10232, and lane runs on it no-op (run 41293781163).
Issue Triage session, amending this issue's Decisions line "REVIEW.md syncs from melodic-software/standards, so the change there is another-repository work (owner stop)": the standards change is agent work, not an owner stop. GitHub Support ruled it 2026-11-01 ~15:5xZ; no owner words or owner-decisions row back the "owner stop" note. Automation Lanes files the melodic-software/standards PR (one repository-neutral REVIEW.md pointer to a reviewed repository's docs/review/README.md); its scope is in the #10231 record. The body is left as filed.
Activity
github-actions commentedon Oct 16, 2026
Proposed work class: work-class: structural
Basis: automated intake triage; the lanes App applied this class.
Size: Large
Complexity: Medium
Uncertainty: Known
Flags: none
Blocked by: none
#7583 (comment)
The decision settles the open questions: standards go into per-concern files, AGENTS.md keeps one pointer line, and the issue closes through its sub-issues (per-concern-split, review-md). Nothing is left to ask.
I found no open duplicates. The dependency #7571 isn't in the open-issue list I was given, so I couldn't confirm its state. Part of the work is in melodic-software/standards, which is a separate repository.
Automated triage, applied by the lanes App.
kyle-sexton commentedon Oct 16, 2026
What the article adds to the standards layout, and the owner's 2026-10-16 decisions. Source: https://claude.dev/blog/solving-code-review-at-anthropic (2026-10-15); ids are digest unit and claim.
Owner decisions, 2026-10-16.
/code-reviewreads CLAUDE.md, and its docs say of REVIEW.md: "It doesn't read REVIEW.md" (https://code.claude.com/docs/en/code-review-local). Without a pointer, a local review sees none of this repository's review rules. The line is a pointer, not a rule, so Q150's "no review rules inside AGENTS.md" holds.## Code Review Rulesmoves into the per-concern files, leaving only the pointer line.The gap is narrow: under Q500 (GUIDE 75) a repository with a CI lane runs no local review, and the CI lanes follow REVIEW.md citations (GUIDE 122). It affects ad hoc local sessions and consumers set to
review_venue: local.A suggestion from the ADR review consensus, not an owner decision: load the per-concern files by path scope, so each enters context only when a file it covers is touched.
This thread already puts the nit cap in its per-concern file (Q570). The article's REVIEW.md traits it lacks (04 C18-C20), which belong in those files:
The article also checks separately whether the reviewer follows REVIEW.md (04 C17).
Background only: managed Code Review is not used here (GUIDE 122, Q594). Q570's generated REVIEW.md stays an opt-in for consumers on that service.
Against the originating sources. This thread rejects Matt's single standards file and Lauren's rule table in the instruction file. The article agrees with pointing out of a short CLAUDE.md (04 C14); its reviewer also flags code that breaks a CLAUDE.md convention (04 C15). The pointer line meets 04 C15 without putting rules in the instruction file.
Related: #7576 (PR template), #7573 (codify destinations), #7638 (whether rules take effect).
kyle-sexton commentedon Oct 24, 2026
Released for re-triage under the owner's 2026-10-24 rule: an issue stops for a person only for an intent or vision question, a security call, a host or attended run, or spend outside prepaid credit. The hold label came off and intake triage runs again now; it re-applies a hold, with a numbered question, when one of those applies.
To hold this issue again, add
needs-human(orstatus: needs-decision) and say why in a comment.13 remaining items
melodic-automation-lanes commentedon Oct 26, 2026
Split at intake (owner Q93). This issue is Size Large, so its done-when is split into 2 sub-issues, each holding its task-list lines as written.
Automated split by the lanes App. Admission waits on this issue while a sub-issue is open; work-unblock closes it as completed once every sub-issue is closed by a merge or a listed human.
kyle-sexton commentedon Nov 1, 2026
Issue Triage session: removing agent-ready from this split parent; it waits on its open sub-issues #10231 and #10232, and lane runs on it no-op (run 41293781163).
kyle-sexton commentedon Nov 1, 2026
Issue Triage session, amending this issue's Decisions line "REVIEW.md syncs from melodic-software/standards, so the change there is another-repository work (owner stop)": the standards change is agent work, not an owner stop. GitHub Support ruled it 2026-11-01 ~15:5xZ; no owner words or owner-decisions row back the "owner stop" note. Automation Lanes files the melodic-software/standards PR (one repository-neutral REVIEW.md pointer to a reviewed repository's docs/review/README.md); its scope is in the #10231 record. The body is left as filed.
melodic-automation-lanes commentedon Nov 2, 2026
Every sub-issue of this split is closed as completed by a listed human or a merge (#10231, #10232, #12198), so the lanes App closes it as completed.