Repository navigation
fix: 0.13 release branch, doc lint and branch-level test reds (#37171, #37451, #37530, #37539) - #37633
Conversation
…on-head) The Source Gates "Lint documentation trees" step fails on release/0.13.0 itself, so every 0.13 backport PR goes red on it. - Link 2026-11-13-the-crew-mode-feature-is-named-conductor-mode.md from docs/decisions/README.md with main's index row. - Bring docs/request-for-change/rfc-kas-session-head-in-agent-prompt.md from main (#36964) and its README index row. session_head.py, its test and three docs on the branch already cite it. Docs only; no code changes. scripts/docs-lint.sh passes on the result.
First Principles Review (Opus 5.5) — ⏭️ skippedRevision |
UX Review (Opus 5.5) — ⏭️ skippedRevision |
Design Review (Opus 5.5) — ✅ PASSDesign-level review of Design-Verdict: PASS Nothing blocks: this is a docs-and-tests-only backport of main's end state, and the private symbols the new tests read are on the branch. Notes
[DESIGN-REVIEWED] 0e11682 Description read[DESCRIPTION-READ] 6101cf436bae9f14dfd857b379f4211a375e74379c32d5477875167b76322e53 That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with |
GPT 6.1 Review — ✅ no blocking findingsGPT 6.1 completed its review of This comment is updated in place on each push. Review detailsNo findings. False positive or not applicable? A repository writer can comment: |
Opus 5.5 Review — ✅ no blocking findingsReviewed Verdict parsed from the review's SHA-scoped output markers for commit False positive or not applicable? A repository writer can comment: |
…#37539 Two backend tests on release/0.13.0 were red for reasons main has already fixed in test-only commits: - test/tracing/test_tracing_otlp.py expected four spans, but #37005 (already on the branch) adds kirocrew.step. Cherry-picked #37171 unchanged. - TestSelfModuleIndexIsLinear failed three runs in a row on the 0.5 s budget (0.522 s, 0.511 s, 0.509 s). Took main's end state of test/test_denied_commands_security_cost.py and test/denied_commands_security_helpers.py, which carries #37451 (min of five readings), #37530 (tracer paused) and #37539 (count work, not CPU time), plus the #37539 hunk of docs/system-specs/common/testing-conventions.md. The picks conflict one by one because the file also carries #28293, a test-only change. No src/ changes. (cherry picked from commit 6de8528)
6b8dbcd to
0e11682
Compare
Problem / Motivation
Goal: The Source Gates "Lint documentation trees" step passes on
release/0.13.0again, so 0.13 backport PRs are not red for a reason they did not cause../scripts/docs-lint.shfails onrelease/0.13.0itself (basede4155bd70, run 43709803472) with three FAIL groups:docs/decisions/2026-11-13-the-crew-mode-feature-is-named-conductor-mode.mdsrc/kiro_crew/session_head.py:8andtest/test_session_head.py:3citedocs/request-for-change/rfc-kas-session-head-in-agent-prompt.mddocs/architecture/context-management.md,docs/system-specs/modules/memory-skills-hooks.mdanddocs/system-specs/modules/providers.mdWhy it matters
Source Gates is a blocking job. While the release branch fails it, every 0.13 backport PR (#37598, #37551, #37553) is red, and a real doc-lint regression in a backport cannot be told apart from this one.
Not a goal
session_head.pyand its test stay as they are on the branch.What changed (motivation → approach → change)
Symptom: the doc lint fails on the release branch. Root cause: backports brought the decision doc and the code citing the RFC, but not the decision index row or the RFC file. Change: take both from main as they are there.
docs/decisions/README.md: add main's index row for the 2026-11-13 conductor-mode name decision, in the same place main has it.docs/request-for-change/rfc-kas-session-head-in-agent-prompt.md: copied unchanged from main (it came in with docs(rfc): send a new chat's session-stable head as the KAS agent prompt #36964).docs/request-for-change/README.md: add main's index row for that RFC, at the top of the table as on main.Once Source Gates passed, the backend shards showed a second branch-level red:
test/tracing/test_tracing_otlp.pyfails on barerelease/0.13.0too (2 failed, 91 passed locally onde4155bd70). The branch has #37005, which adds akirocrew.stepspan, but not #37171, the test update that expects it. The second commit cherry-picks #37171 unchanged (test only, +8/-2).The same commit also fixes the third branch-level red.
TestSelfModuleIndexIsLinear::test_the_scan_is_linear_not_quadraticintest/test_denied_commands_security_cost.pyfailed three runs in a row on its 0.5 s budget (0.522 s, 0.511 s, 0.509 s) and passes locally. Main fixed this flake in three test-only commits: #37451 (hold the budget to the minimum of five readings), #37530 (read it with the tracer paused) and #37539 (count work, not CPU time). Those picks conflict one by one here because main's file also carries #28293 (test-only), so the commit takes main's end state oftest/test_denied_commands_security_cost.pyandtest/denied_commands_security_helpers.py(the helpers diff equals #37539's), plus the #37539 hunk ofdocs/system-specs/common/testing-conventions.md. Nosrc/change; everykiro_crewsymbol the two files import exists on this branch. Folded into the second commit to keep the PR at two commits.Backwards compatibility
Compatible: docs and test files only; no
src/change, so nothing that works today stops working.Removes nothing: the diff adds two index rows and one doc file, widens two span assertions to the span #37005 already emits, and replaces wall-clock budgets in the deny-gate cost tests with main's work-count checks; no product behaviour is taken away.
Tests
Proof: not provable -- test-only change
./scripts/docs-lint.shon the final tree prints "All documentation checks passed" with 0 FAIL lines (before: 3 FAIL groups).python3 scripts/docs_lint.py --testpasses.test/tracing/test_tracing_otlp.py: 2 failed on the bare base, 93 passed with the #37171 pick.test/test_denied_commands_security_cost.py: 111 passed in each of 3 local runs. The 13 other test files that importdenied_commands_security_helpers: 2296 passed, 1 skipped.Manual verification
N/A -- the doc linter run above is the check this PR fixes; CI runs the same script.
Related Issues
Unblocks the Source Gates job on #37598, #37551 and #37553. Backports #37171, #37451, #37530 and #37539.
Sibling sweep
Swept: ./scripts/docs-lint.sh on release/0.13.0 (lists every unindexed doc and every missing cited doc path)
Sibling: docs/decisions/README.md -- changed here
Sibling: docs/request-for-change/README.md -- changed here
Sibling: test/tracing/test_tracing_otlp.py -- changed here (the other branch-level red, found by the backend shards)
Sibling: test/test_denied_commands_security_cost.py -- changed here (timing flake red three runs in a row on this branch)
Pattern harvest
Not generalizable: a backport carried code that cites a doc without the doc; the existing doc lint already catches this class, which is how it was found.
Checklist
feat|fix|docs|style|refactor|perf|test|chore|ci|build|revert: ...)Contribution License Agreement
Contributions are accepted under the project's Apache-2.0 license; see the Licensing section of CONTRIBUTING.md.