Repository navigation
test: name mimocode in the reply-format harness table - #36315
Conversation
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 to check. It is one test-table row that records how mimocode actually behaves: it is not in the set of harnesses whose compaction Kiro Crew sees, so it maps to False. [DESIGN-REVIEWED] 6833128 Description read[DESCRIPTION-READ] 0d8a2f7481c9f15083ec3f44db4fae5b627b7e6a9f611f0de82fff5b4014d284 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 |
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: |
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: |
|
This PR and #36290 make the same change to fix test_the_table_names_every_selectable_harness on main. Both add the one line "mimocode": False to COMPACTION_SEEN in test/test_reply_format_once.py, and the resulting file is identical. Please keep one of #36290 and #36315 and close the other, carrying over anything only this one has. |
Code change approver recommendation: ApproveWhy: the decider estimates a 3% chance that this change introduces a defect, within the 12% limit for approving on its own, and every check passes. Test-only change that adds "mimocode": False to COMPACTION_SEEN. It matches the source: mimocode is in BASELINE_SELECTABLE_BACKENDS (backends.py:666-680), and both providers report compaction only for backends in ACP_BACKENDS_COMPACTION_REPORTED, which is {kiro, kas, codex, claude}. The table/set equality test now passes, and the new parametrized cases expect the block on every turn, which matches reports_compaction being False. All Linux backend shards passed, including shard 8. The two failures are an API rate limit (Analyze actions) and the still-pending readiness signal. An automated recommendation, not a review: a maintainer still decides. It approves on its own only when the estimated chance of a defect is at most 12% and the change has at most 400 changed lines. Decider v3c (217fe607) · commit |
Problem / Motivation
Goal: Keep main green by making the reply-format harness table name every selectable harness.
test_the_table_names_every_selectable_harness fails on main: the table in test/test_reply_format_once.py lacks mimocode, which became selectable a few minutes after the table was written. Shard 8 fails on Linux and Windows for every PR merged with main.
Why it matters
Every open PR that merges with main reds on shard 8 until this lands.
Not a goal
What changed (motivation → approach → change)
mimocode is not in the set of harnesses whose compaction Kiro Crew sees, so its reports_compaction is False and the block is sent on every turn. The table now records that: mimocode: False.
Backwards compatibility
Removes nothing: one test-table entry is added.
Tests
test/test_reply_format_once.py: 88 passed locally, including the table test and the new per-harness cases for mimocode.
Proof: not provable -- test-only change
Manual verification
N/A — unit coverage sufficient; the change is one test-table row.
Related Issues
Refs #35756
Sibling sweep
Siblings: none -- the only table keyed by the full selectable set is this one.
Pattern harvest
Not generalizable: a harness onboarded while another PR froze the selectable set in a test.
Checklist