feat: delete the legacy 17-row eval matrix — the sliced lane is the only paid lane

Every PR paid twice: the hand-enumerated matrix (18 test files, 22.6 min,
~$21 API measured on run 33263204465) ran serialized AHEAD of the strictly
superior sliced lane via 'needs: evals' — 35.5 min wall and ~2x paid spend
for the same diff. 14 of 17 rows carried no tier:, so periodic Opus
benchmarks leaked into every PR (the e2e-plan row alone: 12/12 tests,
21.7 min, $7.28 — the wall-clock bound of ALL of CI).

Parity receipt (static, pre-deletion): the sliced lane's gate census (49
files, derived from the runner itself) strictly contains all 18 matrix test
files, plus 31 files the matrix never ran. Pure deletion — one revert
restores it. The PR comment moved into slices-report (same '## E2E Evals'
upsert marker, now sourced from slice artifacts + carrying the fail-closed
reconciliation verdict). plan-slices loses the needs edge; the dead
workflow-level EVALS_TIER env goes with it.

test/evals-workflow-matrix.test.ts (and its KNOWN_MATRIX_GAPS /
KNOWN_TIER_UNSET burn-down ratchets — retired: the sliced census makes
'every gate file runs' true by construction) is rewritten as
test/evals-workflow-wiring.test.ts: matrix stays deleted, planner/executor/
report tier + slice-count agreement, both surviving lanes on the shared
register-skills composite with its fail-fast verification loop, PR comment
survival. Expected: PR eval wall 35.5 -> ~13 min, per-PR paid spend ~halved.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
Garry Tan
2026-08-31 04:20:40 +00:00
co-authored by Claude Fable 5
parent 0f26859bd1
commit 07c2452e6d
7 changed files with 266 additions and 630 deletions
+4 -3
View File
@@ -7,8 +7,9 @@
* execute anywhere — forever, silently. Four files were in that state
* (codex-e2e-plan-format, codex-e2e-recommendation-substance,
* llm-judge-recommendation, carve-section-loading), and the tripwire built
* for the adjacent class (test/evals-workflow-matrix.test.ts) couldn't see
* them because it filters on isPaidTestFile() FIRST.
* for the adjacent class (the since-retired evals-workflow-matrix test;
* successor: test/evals-workflow-wiring.test.ts) couldn't see them because
* it filtered on isPaidTestFile() FIRST.
*
* Detection is over source text, so meta-tests and helpers that mention the
* gate patterns need reasoned exemptions (same convention as
@@ -30,7 +31,7 @@ const SCANNER_EXEMPT = new Map<string, string>([
// Meta-tests that quote gate-pattern strings to test classification:
['test/helpers/e2e-gate.unit.test.ts', 'free unit test OF the gate predicates (env stubbed)'],
['test/paid-shards.test.ts', 'quotes tier-guard strings as classification fixtures'],
['test/evals-workflow-matrix.test.ts', 'parses tier guards out of matrix files'],
['test/evals-workflow-wiring.test.ts', 'pins the sliced-lane yml wiring (successor to the matrix test)'],
['test/e2e-tier-alignment.test.ts', 'parses tier guards to enforce alignment'],
['test/paid-orphan-tripwire.test.ts', 'this scanner'],
]);