Skip to content

[🔥AUDIT🔥] review: bump installed reviewer to review-v1.13.0; enable blocking-only re-reviews #340

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
jwbron merged 1 commit into from
Aug 11, 2026
Merged

[🔥AUDIT🔥] review: bump installed reviewer to review-v1.13.0; enable blocking-only re-reviews #340

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 6 additions & 1 deletion .github/aw/review/ROUTING
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,12 @@ enable holistic,completeness,test-adequacy,first-principles,conventions,document
# whole-change reviewers, not only the correctness pass. Graduate to
# flip-gated once re-reviews here show scoped is overpaying (see
# workflows/review/README.md for the dial).
re-review scoped
#
# blocking-only: repeat reviews keep the full scoped roster (blocking recall
# unchanged) but post only blocking findings inline; validated non-blocking
# findings collapse to one line each in the review body. First full reviews,
# tripwire re-arms, and guard-degraded full runs still post everything.
re-review scoped blocking-only

@github-actions github-actions Bot Aug 11, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

thought (non-blocking): "Blocking recall unchanged" is vacuous for the advisory-only half of the roster kept on re-reviews. The stated justification for keeping the full scoped roster is that blocking recall is unchanged, but that argument only applies to reviewers that can produce blocking findings (holistic, completeness, test-adequacy, plus the correctness pass). first-principles, conventions, and documentation are advisory-only — checked routing-config.ts: blocking-only 'keeps the configured depth's roster' and only changes the posting surface, and the re-review line offers no per-reviewer dial — so on every re-review those three run at full cost to contribute at most one line each in a

Details fold.

A sketch, not a committable replacement:

If the collapsed one-liners from the advisory trio are judged worth their re-review compute, say so in the ROUTING comment; otherwise consider an upstream lib tweak (skip reviewers whose labels are all non-blocking when blocking-only applies) or fold this into the existing 'graduate to flip-gated' criterion, since blocking-only removes the noise cost but not the compute cost that criterion was watching for.


# Shipped composite actions run inside consuming repos' CI with their credentials.
actions/** tier=high
Expand Down
34 changes: 17 additions & 17 deletions .github/workflows/review.lock.yml
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

11 changes: 7 additions & 4 deletions .github/workflows/review.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -244,7 +244,7 @@ pre-agent-steps:
# `source:` below, so the prompt and the lib it invokes come from one version.
# Even though this IS Khan/actions, the reviewer runs the released lib, not
# the PR head; a PR must not be able to change the code that reviews it.
ref: review-v1.12.0
ref: review-v1.13.0
path: gh-aw-review-lib
persist-credentials: false

Expand Down Expand Up @@ -336,7 +336,7 @@ max-daily-ai-credits: -1
max-ai-credits: 2500
env:
REVIEW_MAX_AI_CREDITS: "2500"
source: Khan/actions/workflows/review/review.md@review-v1.12.0
source: Khan/actions/workflows/review/review.md@review-v1.13.0
---

# PR Reviewer
Expand Down Expand Up @@ -1574,8 +1574,11 @@ proximity:
wrong, even when one calls it a wrong cap, one quotes the comment, and one cites a
doc-comment convention: one rewritten comment satisfies all three.
- **Group** across lines. A defect is routinely flagged at different anchors (the
function, its doc comment, its test), and the pipeline keeps one anchor. Distance in
the file is not evidence of two defects.
function, its doc comment, the call site below it), and the pipeline keeps one anchor.
Distance in the file is not evidence of two defects. (A twin anchored in another FILE
— e.g. the test in `_test.go` — stays ungrouped under the same-`path` rule below,
however clearly it describes the same defect; the pipeline prices that as a duplicate
comment, not a wrong merge.)
- **Do NOT group** a bug and the missing test for that bug. "The cutoff subtracts months
instead of days" and "no test asserts a stale entry is deleted" cite the same facts and
need two different edits; they are two defects.
Expand Down
Loading