Repository navigation
Conversation
Release headers become '## KiteModels vX.Y.Z YYYY-MM-DD' (the dash on the
0.5.6-0.6.8 headers dropped), section headers '### Added/Changed/Fixed'
('#### Changes' read as Changed). Note text is untouched. bin/release reads
the top '## ' section accordingly.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
1-Bort-1
commented
Oct 6, 2026
1-Bort-1
left a comment
Member
Author
There was a problem hiding this comment.
Independent review (advisory)
Verdict: APPROVE · 0 inline, 0 off the diff
Good
- Every one of the 147 headers in CHANGELOG.md now has the target form: an anchored rg for
^## KiteModels vX.Y.Z YYYY-MM-DD$or^### (Added|Changed|Fixed)$matches all 147 lines that start with#, and 55 + 89 + 3 = 147 agrees with the card's counts - The diff only touches header lines; reading every hunk confirms the note text is unchanged, as the card says
- The bin/release edit is the minimal one the card names: two awk patterns and two messages; a read of the whole script turned up no other
###dependency - The rename from
#### Changesto### Changedunder v0.11.7 is listed in the card, so the one text change in a header is not hidden - The changelog still passes .markdownlint.json: MD024 is
siblings_onlyand MD041 is off, so repeated### Addedunder different releases and the missing H1 are both allowed - Leaving out an empty
## Unreleasedsection is right, because bin/release would otherwise readUnreleasedas the top version - The card points out the coming conflict with #321/#312 and says their notes belong in changelog.d/; that is the right place for that decision
Not good
- v0.6.15, v0.6.10 and v0.6.9 still have notes sitting directly under the release header with no
### Added/Changed/Fixedsection; this is harmless for bin/release, but the changelog is not fully on the one convention the card describes - v0.11.4, v0.11.1 and v0.11.0 list their sections as Changed/Fixed/Added instead of Added/Changed/Fixed; leaving that alone is fine for a headers-only PR, but the card could have said so
- The card says Julia tests were not run and GitHub CI is pending; that is acceptable because no .jl file changed, but CI should be green before merge
claude, rubric CLEAN_CODE.md. A different lab from the implementer
on purpose: a reviewer sharing its blind spots would not flag its mistakes.
Member
Author
|
Local full suite: FAIL (2 min, Julia 1.13.0, one cell of the matrix) |
1-Bort-1
marked this pull request as ready for review
October 6, 2026 12:24
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
TL;DR
Every release header in
CHANGELOG.mdis now## KiteModels vX.Y.Z YYYY-MM-DDand every section header### Added/### Changed/### Fixed, the one convention the box'srules/releasereads. With the old###releases and no##line at all,agent releasefound no version at the top and a bump would have appended its section at the end of the file.What changed
Headers only:
git diff -U0 CHANGELOG.mdhas 0 changed lines that are not headers, so the note text is byte-for-byte what it was.### KiteModels vX.Y.Z …→## KiteModels vX.Y.Z YYYY-MM-DD; the 19 headers from v0.5.6 to v0.6.8 also lose their-before the date.####→###, the one#### Changes(v0.11.7) among them becoming### Changed; the 3 under v0.11.8 and v0.11.6 that already were###stay as they are.rules/releasereads neither.## Unreleasedsection: main has no pending notes, and new ones go inchangelog.d/per the gate.This repo's own
bin/releasekeyed on^###, so after the rewrite it would have taken### Addedfor the top release. Its two awk patterns and two error messages now say##, and nothing else in it moves; bringing the whole script to the box's copy is whatagent releasedoes on the bump.Where I'd push back
PRs #321 and #312 both add a
#### Changedblock under the already-released### KiteModels v0.11.17line. They will conflict on that header line here, and their notes belong inchangelog.d/rather than under a release that has shipped.The install failure the box reported is not this branch's:
bin/install -yinstantiated and precompiled, then ranPkg.test(), which failed on the red KPS4 steady-state test (#314). #326 takes the test run out ofbin/install. Nothing here needs Julia, so I did not wait on it.Verification
rules/release --dry-run --sha origin/main→Version mismatch: Project.toml is 0.11.17 but CHANGELOG top is (expected v0.11.17).rules/release --dry-run --sha 9cb0f9c6(Agents@81d9a5e): exit 0,Version: v0.11.17, notes = the v0.11.17### Addedblock./bin/release --dry-runon the branch: exit 0, same version and notes ·bash -n bin/releasecleanreuseis not installed on the box); passes in GitHub CItest_find_steady_state(test/test-kps4.jl:581,solver returned non-finite valuesafter 247 iterations). This branch did not cause it. It is find_steady_state! does not converge for KPS4 on Julia 1.13, and CI hides it #314: KPS4 has no steady state on Julia 1.13, and main was already red on it before this branch. The branch'ssrc/,test/,Project.tomland manifest are identical to main's, andtest/test-kps4.jlrun alone on this tree gives the same result: 208 passed, 0 failed, 1 errored, 11 broken. The 9 intest_residual!and the 1 intest_gettersare in the Broken column, not Fail. Under GitHub CI theCI=truecatch turns the same error into@test_broken, as find_steady_state! does not converge for KPS4 on Julia 1.13, and CI hides it #314 says.TestOnPRs.ymlruns the Julia tests only onsrc/**,test/**,*.toml)Scope
+148 / -148 across 2 files: 144 header lines in
CHANGELOG.mdand 4 lines inbin/release. Lines out equal lines in because every edit rewrites a line where it stands.Opened by
1-Bort-1, an AI agent working for @1-Bart-1.Closes #328 · task
KiteModels.jl-328