Review round of 0f3bab4 answered in 3f83c15: the finding and the writing hint (README.md:231) are fixed; the thread is resolved.
Same point as 1769, no change: the scenario names the condition by its meaning, Jev judged every changed file; the other inputs (no LLM skip, Jev configured) are preconditions of that judgment,…
Review round of aba8254 answered: both findings repeat 1768/1769 and are intended; resolved without a code change. The ORANGE needs a human decision.
Intended, no change (same point as 1768): the statistics do not under-report a failed call silently.
They count it in unknownCostReviews ("N with unknown cost") and mark the LLM sum as estimate…
Review round of 72e9d8b answered in aba8254: one hint clarified, one intended; both resolved.
Clarified in aba8254: a newer head ends the review with either role; that sentence now stands apart from the decides one.
"Jev judged every changed file" is the definition of triage_complete…
Intended, no change: the key is shared by all reviews of the host, so the difference of the two key limits is no reliable cost of this one review, and werkrevisor does not book it as such. The…
Review round of 8202fe0 answered in 72e9d8b: both findings fixed or clarified and resolved; 538 tests pass.
Clarified in 72e9d8b: the scenario now says that model_failed is YELLOW and becomes a hint only when Jev judged every changed file.
The test text "review exceeded 720 seconds" is the real…
Fixed in 72e9d8b as far as possible: PR-Agent returns no usage for a call that raises, so the cost stays "unknown" (as it did before, when the same failure ended the review as ERROR). The handler…
Review round of 3de4a38 answered in 8202fe0: both findings fixed and resolved; 538 tests pass.
Fixed in 8202fe0: the test now also checks werkator.toml (factor 1.05); it was not in the list before this PR either.
Fixed in 8202fe0: Superseded is re-raised before the catch-all, tested by test_a_superseded_head_still_ends_the_review_when_the_model_only_advises (red without the fix).
Other failures of the…
Fixed in de1a409: a judged path that only Jev's security answer holds is listed under files_touch_security, covered by test_a_build_file_that_touches_security_is_named_as_such.
Answer to the review of 5d7fb3e (YELLOW):
- YELLOW
cli.py:312-317(docs-only PRs and advisory): intended; resolved. - YELLOW
pathjudge.py:154-173(security-held build file named as build…
Not changed, as the user decided: the repository config, read from the default branch and changed only with a human approval, names its few hard paths itself. A host-only pin of paths, e.g.…
Not changed, intended: a docs-only PR has nothing Jev could miss, its PR documents are judged by the classifier, and with model_role = "advises" the model's doubts are hints by design; with the…
Answer to the review of 052d13e (GREEN, merge held for 2 hints): both answered without a change and resolved. Restarting the review to release the merge.