mirror of
https://github.com/garrytan/gstack.git
synced 2026-11-01 06:24:41 +01:00
* chore(scripts): retire hollow and legacy paid package scripts behind one-release stubs (W8a, DX-4, CEO-31) test:codex*, test:evals*, test:e2e*, test:gate, test:periodic, eval:bg, eval:bg:all, eval:flake-rank, eval:watch and test:audit now print their replacement and exit 1 via scripts/retired-command.ts. CONTRIBUTING.md carries the same table; test/retired-commands.test.ts pins both. Eval CLI empty-state hints point at eval:bg:pr. * chore: delete dead update-readme-throughput, eval-watch, orphan fixtures and QA_FUNCTIONAL_INPUTS (W8b, CEO-24) Retired with named behavior: - update-readme-throughput.ts + readme-throughput test: replaced a README anchor that no longer exists from a JSON file that was never produced; the script was a no-op. Guarantee withdrawn (nothing to protect). - eval-watch.ts + its dashboard tests: rendered ~/.gstack-dev/e2e-live.json and _partial-e2e.json, written only by unsharded runs, so it showed nothing for every supported (sharded) entry point. Guarantee withdrawn; live progress is the gstack-detach log or gh run watch. Manual-review rendering stays covered by eval-list and eval-summary tests. - Orphan fixtures (autoplan-pending-artifact-ae.json, autoplan-existing-app/, autoplan-sign-in-ui/, parity baselines v1.47.0.0/v1.57.7.0/v1.69.1.0): no test, touchfile or script reads them; superseded baselines stay in git history. - QA_FUNCTIONAL_INPUTS: exported, never imported. * refactor(test-free): delete dead WORKER_HOSTILE list and buildShardArgs parallel mode Neither had a production caller: full-suite mode runs shard processes, so the worker quarantine was inert and --parallel was never passed (W8b). * docs(test-value): narrow prompt-byte contract to machine-read tokens (D2) Define prompt-byte contract as machine-read tokens (markers, field names, command lines, enum values, file paths, security-boundary strings) and add a sentence-pin glossary entry: convert to structural checks or delete when a behavioral eval covers the behavior. * test(ratchet): count free tests that slice a paid test's source A free test that binds readFileSync(<paid test file>) and cuts it with .slice() tests the paid harness's code out of context even when it imports a lib/ symbol. Record the 15 existing files with reasons; new ones fail with fix text. Self-test covers positive and negative shapes. * test(browse): un-quarantine compare-board; all 16 board tests pass headless (W8d, CEO-24) The GSTACK_COMPARE_BOARD_TESTS opt-in hid a headless mismatch that no longer reproduces: all 16 tests pass on Bun 1.4.0 with the bundled headless Chromium, solo, repeated, and beside other in-process browser files. Protected behavior (the comparison-board DOM contract the plan-design-review/office-hours/design-consultation feedback loop polls) is checked on every free run again. * test: replace silent existsSync early returns with explicit skips or assertions (W8d) - browse/test/build.test.ts: server-node.mjs checks skip only when the bundle is absent and GSTACK_EXPECT_BINARIES is unset; the free CI lane and test:ubicloud build it and set the flag, so a dropped build step fails with the build command instead of passing with no assertions. - gstack-update-check: VERSION is tracked; the guard could only hide a bug. - gen-skill-docs host leakage/frontmatter checks assert the host tree exists (the module-level render always produces it). - cso-docker staged-runtime test uses test.skipIf so reports show it as skipped, not passed. * test(ship): convert ship-workflow-clarity sentence pins to structural checks Add test/helpers/prompt-structure.ts (between, expectTokens, expectAbsent, expectOrdered, expectMentions) with fix text on failure. ship-workflow-clarity now pins headings, step order and machine-read tokens (commands, fields, enum and status strings); safety rules are meaning-level keyword checks. Wording is judged by the 'ship/SKILL.md workflow' judge in skill-llm-eval. Sentence pins: 90 -> 0. * ci: add platform-qualification lane for opt-in ML, gitleaks and Swift tests (W8d, CEO-24) security-bench + security-live-playwright L4 (SECURITY_BENCH), the real gitleaks 8.30.1 memory-ingest cases (GSTACK_TEST_GITLEAKS) and the DebugBridge Swift build invariants (GSTACK_TEST_SWIFT) self-skip in every lane today. No scheduled lane has their prerequisites, so per CEO-24 they move to an owned dispatch + quarterly workflow instead of being deleted. Each job was run locally (ML 10 pass, gitleaks 130 pass; Swift needs macOS). test/platform-qualification-workflow.test.ts pins gate variable, job and file together. * fix(scripts): capture-baseline and preflight-agent-sdk no longer act on --help or typos (W8e) capture-baseline parses its flags strictly: --help prints usage, an unknown or valueless flag exits 2, and only a valid invocation writes a fixture (--help used to write test/fixtures/parity-baseline-current.json). preflight-agent-sdk gains --help, rejects unknown flags, and runs the live SDK query (API spend) only with --live. * test(evals): register the three keyless periodic paid files with case ids (W1b) skill-e2e-auq-consistency, skill-e2e-auq-verbose-vs-carved-ab and codex-e2e-recommendation-substance now carry touchfiles, a periodic tier and kind rule (one trial, unchanged), so editing one selects its own case instead of restoring the full PR gate. The verbose A/B case also registers the vendored pre-cut fixture it reads, and review-army-migration-safety registers the SQL fixture it reads. * feat(evals): narrow PR fallback for free-only inputs (W1a, CEO-13/20, ENG-13) The PR profile now exempts the duration seeds, every free-lane workflow (explicit list; inventory test fails on an unclassified workflow, .yaml and workflow_call included), scripts/ubicloud/** and fixtures declared in the explicit FREE_FIXTURES map. Each remaining unknown file carries a label and fix (needs touchfile entry / add to FREE_ONLY_PR_FILES / real unknown dependency). evals.yml itself still restores the full gate. * ci: delete ci-image.yml; the eval workflows already build and push the same content-hash tag (W8f) evals.yml, evals-periodic.yml and evals-marathon.yml each compute the identical hashFiles() tag and build/push it (with :buildcache and :latest) when missing, so ci-image's push trigger and weekly cron were no-ops (30-day median 0.5 min, 36/36 success). ci-image-tag-binding now pins the three eval workflows. * ci(evals): key periodic/marathon concurrency on ref+event; branch periodic dispatch skips the gate census (W2a, ENG-9, CEO-04) * refactor(paid): drop ~160 unused imports; leaf paid-types/paid-select modules break the paid-* import cycle (W8c) * ci(quality-gate): per-PR concurrency, free runner, drop unscoped bun audit (W2a, W8f) - Concurrency keys on pull_request.number || run_id, so back-to-back main pushes no longer cancel each other's credential scan. - ubuntu-24.04 instead of a billed 8-vCPU Ubicloud runner for a ~1 min job. - bun audit --audit-level=critical ran on every PR regardless of diff, so a newly published advisory turned unrelated PRs red, with its own ignore policy. dependency-review (lockfile PRs, fail >= high) and the weekly OSV sweep remain the dependency gates. * chore(test-free): refresh the free duration seed from the audit's Ubicloud recording Adds the 19 files merged after the last refresh, including the 115 s claude-overlay-setup that packed as ~1 s and doubled a shard (W6a). The seed is re-recorded on the final tree at the end of this wave. * ci: move tiny ubicloud-standard-2 jobs to free ubuntu-24.04 runners (W8f) actionlint, skill-docs, dependency-review, version-gate and pr-title-sync are sub-minute jobs; GitHub-hosted runners are free for this public repo. * feat(test-free): pessimistic unseeded packing, seed growth ratchet and CI health summary - Unseeded files pack at the seed's p99 instead of p75 (W6a). - --ci-plan warns in the job summary past 5 unseeded files; --ci-verify lists shards that outran their prediction by 1.5x (and 30 s) plus every flaky pass absorbed by the serial retry (W6a, W7a job summary). - test/free-seed-ratchet.test.ts fails when a seeded file over 60 s is not on scripts/free-test-seed-allowlist.json, when an entry is stale, and when the allowlist grows versus its merge-base copy unless the new entry is an annotated exception with a todo (CEO-10, DX-7). free-plan runs it with full history so the shrink check is real in the required lane. - Every message names the fix command and docs anchor (DX-6). * feat(evals): state PR coverage mode, counts, reuse and fallback files in the summary and comment (CEO-13) * test: make the tracked-tree skill-doc freshness check local-only; tidy skill-docs header (W8f) Every free CI shard regenerates the tracked SKILL.md files before running tests, so the --dry-run freshness assertion compared the regenerated tree with itself and could not fail there. It now skips under GitHub Actions; skill-docs.yml's git diff --exit-code is CI's freshness gate. Removes a stale concurrency comment in skill-docs.yml. * ci(free-tests): install zsh so the bash+zsh portability arms run (W6f) ~42 zsh-arm tests (aside-probe-shell, qa-browser-preservation, generated-docs-fences) skipped in CI because zsh is not in the runner image. All 141 tests in those files pass with zsh present. * feat(evals): executors write negative receipts at the execution boundary (CEO-25, ENG-6) A completed single-trial rule/judge shard that did not pass now writes its negative receipt beside the pass receipts in the slice artifact, so a cancelled run's FAIL still blocks an older PASS when its report never runs. Receipts merge negatives first and order by (run attempt, completedAt); a same-second FAIL/PASS tie resolves to FAIL. The runner call site is a one-line hunk in scripts/test-paid-shards.ts. * docs(cso): state that contained runtime/scanner execution is unavailable today The scanner catalog is empty while its build inputs await review and the runtime catalog holds no qualified profile until the private evaluator exists, so /cso runs static assessment only. AGENTS.md, README and docs/cso-release-qualification.md now say so; docs/cso-scanners.md already did. * test(helpers): expectMentions splits on sentence ends only Clauses joined by ';' or ':' are one rule; splitting there made a reworded rule fail for punctuation. * test(plan-ceo): convert skill-ceo-section-ordering sentence pins Safety and authority rules (never/do not/stop/only/without/wait/approval) are meaning-level expectMentions checks; other sentence pins are dropped. Headings, labels, option rows, placeholders and machine-read lines (auto-decide annotation, Mode:/approved decisions format, STATUS/SOURCE fields, CROSS-MODEL marker, gstack-ceo-mode-handoff) stay as tokens; ordering checks unchanged. Behavior is covered by the periodic PTY case in skill-e2e-plan-ceo-review-section-loading.test.ts. Sentence pins: 210 -> 23. * test(helpers): expectMentions treats wrapped lines as one sentence Templates hard-wrap prose; only sentence ends and blank lines separate rules. * test(gen-skill-docs): convert sentence pins to meaning checks or drop them Safety/authority sentences become expectMentions keyword checks; other English sentences are dropped (wording is judged by the per-skill workflow judges in skill-llm-eval and the E2E cases in touchfiles-data). Commands, config keys, markers (PLAN MODE EXCEPTION, NO UNRESOLVED DECISIONS, Gate outcome), field names and alternation-based meaning regexes stay. Removed the now-empty 'contains opt-in AskUserQuestion text' test (only pinned a sentence; the resolver's AskUserQuestion wiring is covered by the codex second-opinion tests in the same describe). Lane E's existsSync hunks are untouched. Sentence pins: 216 -> 65 (remaining lines carry machine tokens). * ci(windows): shard the Windows lane via --ci-plan/--ci-run/--ci-verify; move native campaigns to native-qualification.yml - windows-free-tests.yml: a Linux plan job curates and packs the Windows-safe files into 6 shards, one strict serial shard per windows-latest job, and a Linux aggregate verifies exact coverage and every result (8.1 min on one runner today). --windows-only now combines with the CI modes; the plan binds the curated file set (W6e). - Every Windows plan and --list run prints the curated count with its top exclusion reasons in the job summary; a free ratchet fails when the curated subset drops below its floor, with the fix (W6e). - The four Dia/cookie diagnostic dispatch campaigns move to the dispatch-only native-qualification.yml with one exclusive `mode` choice input, so their red dispatch runs no longer read as a red PR lane (W8f). Their workflow pins move to test/windows-native-workflows.test.ts; the boolean-exclusivity guard is replaced by the choice input. * feat(paid): per-slice CI job ceilings and an in-process slice deadline with not_run/hung outcomes and checkpoints (W2c, ENG-2) * fix(pass-rates): segment trial history strictly by EVAL_POLICY version Reader-compatibility floor for EVAL_POLICY v2 (ENG-14/CEO-35/DX-12). Only trials under the reader's own policy version feed labels and alarms; older policy trials stay visible but are never scored or used as a drift baseline; trials from a newer policy are ignored with a printed count and fix text. Reverting the later v2 writer commits leaves a reader that cannot pool v2 records into v1 series (the pre-PR reader at2db0b3adoes: it reports a mixed fixture as FAILING with drift alarms). * docs: one contributor command table; retired commands, seed sources and current workflows (W8g, DX-5, DX-13, CEO-12) - CONTRIBUTING.md gains 'Which command do I run?' (task -> command -> cost/time -> needs, including branch validation via gh workflow run); CLAUDE.md, AGENTS.md and TESTING_INTERNALS.md link to it instead of repeating command lists. Tier tables drop retired commands and stale per-run dollar estimates. - Retired names (test:evals, test:e2e, eval:bg, eval:flake-rank, eval:watch, ...) are gone from CLAUDE.md, CONTRIBUTING.md, TESTING_INTERNALS.md, ARCHITECTURE.md and USING_GBRAIN_WITH_GSTACK.md. - The flake-ledger claim now says what happens: CI uploads flake-ledger-* artifacts; eval:pass-rates reads only the local ledger. - TESTING_INTERNALS states that only test:ubicloud --record-durations produces an acceptable duration seed (DX-13). - CI section: quality gate no longer runs bun audit; dependency-review's real trigger; the platform-qualification lane. - docs/test-audit-2026-09.md gets a corrections section for its stale claims. - TODOS.md: audit follow-ups (post-merge metrics, stub removal, name reclaim, diff-aware planner); eval-watch shard-awareness and the image freshness tripwire closed as won't-do. * feat(evals): recover receipts of cancelled PR runs (W1c, CEO-14/34, ENG-4/5/7) A new recover-receipts job runs base-ref code with actions: read, restores the PR's store, merges the receipts of completed cancelled evals.yml runs for this PR (provenance-checked, at most 5, oldest first, negatives first), reconstructs completed panel negatives with panelVerdict() (partial panels get no credit) and keeps a recovery.json checkpoint that never advances past an unresolved run. plan-slices keeps contents: read, takes the store as an artifact, turns reuse off when recovery degraded or a run is unresolved, and carries the whole store to slices-report so nothing is lost. Any gh problem degrades to no recovered receipts with a named reason. * feat(evals): classify this wave's free-only tools and workflows for PR selection platform-qualification.yml, retired-command.ts, test-health-report.ts, bump-harness-version.ts and eval-bg.ts never run inside a paid case. * feat(evals): debounce PR push bursts before planning paid slices (W1d, CEO-15, ENG-11, DX-7) A tiny ubuntu-24.04 job waits 90 s only on synchronize when another run for the PR started in the last 15 minutes; cancel-in-progress supersedes, and a run that outlives the wait reports "superseded by <sha>" and plans nothing. Labels: evals-no-debounce skips the wait, evals-fresh disables receipt reuse. * perf(test): move the real-process browser settlement cases out of test-free-shards.test.ts (W6b) The 19 owned-browser settlement modes were 85 s of the file's 95 s and could not be packed apart from the fast runner unit cases. They move to a shared fixture with two files (lifecycle 47 s, ownership 39 s); the unit file drops to 10 s. Same modes, same assertions. * chore(evals): delete the finished validation_phase dispatch and its inline planner (W8f) Retirement (CEO-24): the protected behavior was a one-off cookie-repair validation subset (eight dependent cases) and a quality/behavior split of the census; that validation is complete, so the guarantee is withdrawn. Branch validation dispatches evals.yml with evals_all. * perf(test): inject the office-hours record grace in non-cooperative deadline tests * perf(test): shrink aside-render process slack via GSTACK_RENDER_SLACK_MS in timeout tests * perf(test): make the memorable lock give-up configurable for lock-contention tests * perf(test): check install-timeout knob normalization through its own setup lines * perf(test): inject a 2 s storage-reset deadline in the late isolated dispatch test * perf(test): shorten the slow gbrain fake to 1 s, still past the 300 ms probe timeout * ci(free-tests): run only cso-docker-integration in the free CSO Docker job (F16) cso-node-lifecycle-integration and cso-stack-cold-integration need GSTACK_CSO_TEST_STACK images and always skipped there; they stay in the CSO runtime workflows through test:cso:docker. * test(gen-skill-docs,plan-ceo): split mixed sentence pins into tokens and rules Sentences that wrap a backticked or bold token keep the token (expectTokens) and, when they state a safety rule, a keyword check; short safety sentences (25-39 chars) get the same treatment. * test(plan-ceo): convert ceo-workflow-clarity.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: periodic PTY case skill-e2e-plan-ceo-review-section-loading and the plan-ceo workflow judges. Sentence pins: 13 -> 1. * test(plan-eng): convert eng-review-routing.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: plan-eng E2E cases (skill-e2e-plan-eng-*) and the plan-eng workflow judge. Sentence pins: 28 -> 6. * test(plan-eng): convert eng-scope-entry-ap.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: plan-eng E2E cases (skill-e2e-plan-eng-*) and the plan-eng workflow judge. Sentence pins: 16 -> 5. * test(plan): convert plan-scope-selection.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: plan-review E2E cases and workflow judges in skill-llm-eval. Sentence pins: 19 -> 2. * test(qa): convert qa-lazy-sections.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: qa/qa-only E2E cases (skill-e2e-qa-*) and the qa workflow judge. Sentence pins: 71 -> 11. * test(qa): convert qa-browser-preservation.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: qa/qa-only E2E cases (skill-e2e-qa-*) and the qa workflow judge. Sentence pins: 25 -> 11. * test(qa): convert qa-probe-gates.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: qa/qa-only E2E cases (skill-e2e-qa-*) and the qa workflow judge. Sentence pins: 20 -> 5. * test(qa): convert qa-checkpoint-evidence.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: qa/qa-only E2E cases (skill-e2e-qa-*) and the qa workflow judge. Sentence pins: 7 -> 5. * test(qa): convert qa-evidence.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: qa/qa-only E2E cases (skill-e2e-qa-*) and the qa workflow judge. Sentence pins: 4 -> 3. * test(design): convert design-consultation-contract.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: design-consultation E2E/judge cases in touchfiles-data. Sentence pins: 44 -> 11. * test(outside-voice): convert run-in-background-guidance.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: outside-voice E2E cases and plan-review workflow judges. Sentence pins: 35 -> 5. * test(spec): convert spec-template-invariants.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: spec E2E/judge cases in touchfiles-data. Sentence pins: 36 -> 21. * perf(evals): 420 s census slice budget; case-shard skill-e2e-qa-workflow; seed its case walls (W5c) * feat(evals): evals.yml dispatch contract for eval:bg (CEO-17/29, DX-1, ENG-8/12) workflow_dispatch gains base_ref, base_sha, expected_sha, nonce and pr_receipts. The nonce appears in run-name so the dispatcher resolves its own run; evals_all=false runs the PR profile against base_sha/base_ref; a checkout that differs from expected_sha fails before any paid work with a named fix; pr_receipts reads that PR's receipts read-only through the recover-receipts artifact path (never the cache) and never saves. * ci(windows): pack the Windows lane with Windows-measured durations (W6e) The first sharded run (37221451711) finished in 6.0 min: the Linux seed has browse/test/cookie-import-native-job.test.ts at ~0 s (its cases skip on Linux) but it takes 122 s on Windows, so one shard ran 242 s against an 84 s prediction. The plan now reads scripts/free-test-durations-windows.json; a record_durations dispatch times every Windows-safe file alone and uploads it. * test(free): settlement case files import the runner rule they exercise Keeps the split files out of the test-of-test ratchet: the retry-eligibility rule is a direct runner import passed to the shared fixture. * fix(marathon): no hollow plan-decision-classification shard; office-hours marathon scenario scoped to the pilot (W2f) * test(free): satisfy the typecheck:test ratchet in the seed ratchet test * refactor(ci-history): one shared GitHub history reader with the CI flake-ledger reader Extract pass-rates' gh fetch layer into scripts/lib/ci-history.ts (CEO-09) and add the W7a reader: paged run listings that fail whole on any page error, per-run artifact downloads (cached, size-capped, data only), flake-ledger parsing, the pre-registered flaky-file rule (> 5% of main free-tests runs over a window of at least 20 runs, widened to the newest 20 runs when the window is thinner), a job-summary table, and named unavailable reasons with next steps. * feat(test-health): test:health report of the audit success metrics scripts/test-health-report.ts (CEO-09/06/22/32, DX-8/9/12/13) prints every success metric with its counting rule from CI history: PR full-fallback rate, reused PR trial records, PR-lane spend per merged PR and per validated revision, cancelled slice-minutes per week, push-to-verdict p50/p95, census expected failed trials per run with raw failure rate per kind/class and P(lane red) under the recorded panels vs single-trial aggregation, INFRA re-dispatches, quarantine evidence (insufficient evidence distinct from healthy), free flaky-pass rate, free/Windows wall, free serial time and the net-complexity line. An unreadable input prints 'unavailable: <reason>; next: <step>', never 0. --enforce fails on a flaky file (W7a) or > 5 unseeded free files (CEO-10); unavailable checks never fail and keep the issue open. * test(review): convert review-workflow-clarity.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: review E2E cases (skill-e2e-review*) and the review workflow judge in skill-llm-eval. Sentence pins: 85 -> 17. * test(review): convert review-entry-and-design-clarity-au.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: review E2E cases (skill-e2e-review*) and the review workflow judge in skill-llm-eval. Sentence pins: 15 -> 4. * test(review): convert review-enum-lifecycle.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: review E2E cases (skill-e2e-review*) and the review workflow judge in skill-llm-eval. Sentence pins: 8 -> 0. * test(ship): convert ship-control-flow.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: ship E2E cases (skill-e2e-ship*, skill-e2e-workflow) and the 'ship/SKILL.md workflow' judge. Sentence pins: 74 -> 9. * test(ship): convert ship-skip-requeue.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: ship E2E cases (skill-e2e-ship*, skill-e2e-workflow) and the 'ship/SKILL.md workflow' judge. Sentence pins: 28 -> 3. * test(ship): convert ship-plan-completion-invariants.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: ship E2E cases (skill-e2e-ship*, skill-e2e-workflow) and the 'ship/SKILL.md workflow' judge. Sentence pins: 28 -> 12. * test(ship): convert ship-reentry-gates.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: ship E2E cases (skill-e2e-ship*, skill-e2e-workflow) and the 'ship/SKILL.md workflow' judge. Sentence pins: 16 -> 4. * test(ship): convert ship-document-release-dispatch.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: ship E2E cases (skill-e2e-ship*, skill-e2e-workflow) and the 'ship/SKILL.md workflow' judge. Sentence pins: 11 -> 4. * test(ship): convert ship-apple-gate.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: ship E2E cases (skill-e2e-ship*, skill-e2e-workflow) and the 'ship/SKILL.md workflow' judge. Sentence pins: 7 -> 0. * test(ship): convert ship-skip-actor.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: ship E2E cases (skill-e2e-ship*, skill-e2e-workflow) and the 'ship/SKILL.md workflow' judge. Sentence pins: 6 -> 1. * test(ship): convert ship-template-redaction.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: ship E2E cases (skill-e2e-ship*, skill-e2e-workflow) and the 'ship/SKILL.md workflow' judge. Sentence pins: 5 -> 4. * test(ship): convert ship-publication-gates.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: ship E2E cases (skill-e2e-ship*, skill-e2e-workflow) and the 'ship/SKILL.md workflow' judge. Sentence pins: 5 -> 2. * test(ship): convert ship-review-loop.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: ship E2E cases (skill-e2e-ship*, skill-e2e-workflow) and the 'ship/SKILL.md workflow' judge. Sentence pins: 3 -> 1. * test(ship): convert ship-version-sync.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: ship E2E cases (skill-e2e-ship*, skill-e2e-workflow) and the 'ship/SKILL.md workflow' judge. Sentence pins: 5 -> 2. * test(ship): convert ship-pr-liveness-policy.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: ship E2E cases (skill-e2e-ship*, skill-e2e-workflow) and the 'ship/SKILL.md workflow' judge. Sentence pins: 4 -> 0. * test(plan-review): convert plan-review-cases.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: plan-review E2E cases and workflow judges in skill-llm-eval. Sentence pins: 58 -> 15. * test(autoplan): convert autoplan-phase-order.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: autoplan E2E cases (skill-e2e-autoplan-*) and the autoplan workflow judge. Sentence pins: 42 -> 11. * test(autoplan): convert autoplan-phase-handoff.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: autoplan E2E cases (skill-e2e-autoplan-*) and the autoplan workflow judge. Sentence pins: 20 -> 4. * test(autoplan): convert autoplan-snapshot.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: autoplan E2E cases (skill-e2e-autoplan-*) and the autoplan workflow judge. Sentence pins: 21 -> 5. * test(outside-voice): convert outside-voice-preflight.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: outside-voice E2E cases and plan-review workflow judges. Sentence pins: 26 -> 9. * test(ask-format): convert resolver-ask-user-format.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: AUQ E2E cases (skill-e2e-auq-*); prompt-injection rule kept as a meaning check. Sentence pins: 32 -> 15. * test(skills): convert binding-template-drift.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: workflow judges in skill-llm-eval. Sentence pins: 19 -> 13. * test(skills): convert carve-plan-fixture.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: workflow judges in skill-llm-eval. Sentence pins: 20 -> 6. * test(land-and-deploy): convert land-and-deploy-postfail.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: land-and-deploy E2E cases and its workflow judge. Sentence pins: 21 -> 14. * test(land-and-deploy): convert land-and-deploy-ci-gate.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: land-and-deploy E2E cases and its workflow judge. Sentence pins: 5 -> 4. * test(land-and-deploy): convert land-and-deploy-flow.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: land-and-deploy E2E cases and its workflow judge. Sentence pins: 5 -> 3. * test(office-hours): convert office-hours-completion.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: office-hours E2E cases (skill-e2e-office-hours*). Sentence pins: 16 -> 1. * test(plan-ceo): restore the stale closing-route guard The negative check for 'Closing hooks' / 'return to section' guards a retired route token, not a sentence; the first conversion pass dropped it. * ci(test-health): weekly test-health workflow with an honest tracking issue New test-health.yml (CEO-18/DX-9): Monday 10:00 UTC after the periodic census plus dispatch, free ubuntu-24.04 runner, permissions contents:read + actions:read + issues:write only. Runs bun run test:health --since-days 7 --enforce; an enforced failure upserts one tracking issue, a run with unavailable inputs keeps it open and lists them, and it closes only when every enforced check has evidence and passes. Only main runs touch the issue. * feat(evals): an edited paid test file runs its own gate cases in the PR lane (DX-11) A gate case outside the audited fast profile is selected when its single owning paid test file changed and the runner can select it by name (its Bun test is named by the case id). The manifest records directCases and validation accepts only recorded, name-addressable ones; the runner and planner read the file->cases map from test-pr-profile. Cases the runner cannot address stay deferred with the fix. An inventory test covers every owned gate case. Touches one hunk each in test-paid-shards.ts and lib/paid-plan.ts, and pr-shared-input-selection.test.ts's ship-skip pin. * ci(osv): tracking issue for scheduled reds and a 14-day suppression expiry test F11/W8f: a failed main OSV scan now upserts one tracking issue and a clean scan closes it (scheduled reds notified no one during the August backlog). A free test fails 14 days before any .osv-scanner.toml ignoreUntil (all four expire 2026-11-30), with the fix: upgrade and delete, or re-justify and move the date. * test(ship,review): keep guards against retired fail-open instructions 'skip silently' on reviewer failure and 'do not block /ship on subagent failure' were harmful instructions, not prose style; keep them absent with the shortest distinctive phrase. * feat(eval-select): default to the pr profile DX-10: eval:select previews the PR lane's selection by default, matching test:pr and eval:bg:pr; --profile full (or EVALS_PROFILE=full) keeps the plain touchfile selection. * test(evals): type-clean the recovery gh fake (typecheck:test ratchet) * chore(windows): record the Windows duration seed (run 37221963236) 573 Windows-safe files timed alone on windows-latest; cookie-import-native-job is 102 s there. Six shards now predict 154 s each. * feat(evals): eval:bg dispatches the pushed ref to CI or runs locally under a 1.5x-planned cap (W5d, CEO-17/29, DX-1/2/3/7, ENG-12/19) * feat(runners): --help on both shard runners via the shared flag loop; paid runner header describes today's planner/executor/report (W8e) * test(windows): drop the stale Playwright exclusion and force-include probe-proven files (W6e) A temporary windows-latest probe (run 37221995077) ran each of the 97 files excluded for 'launches Playwright browser' or 'raw /tmp/ path' alone: all 17 Playwright files pass (Chromium is installed and Bun drives it there), and 41 of the /tmp files pass with executed cases. The Playwright pattern is removed and the 41 files are force-included with the probe as their reason; the 36 real failures stay excluded. The curated subset grows from 573 to 631 files and the ratchet floor follows. Windows curation moves to scripts/lib/windows-curation.ts (the runner drops to 2000 lines). * test(free): one install fixture cloned from a read-only seed (W6c, ENG-1) Merge test/helpers/setup-codex-scope-fixture.ts into install-fixture.ts. The checkout is built once per test file into a read-only seed and cloned per fixture with copy-on-write where the filesystem has it (full copy otherwise), restoring each file's own mode; clones never hard-link the seed because put() and tests write in place. Callers register afterAll(cleanupSeed); Bun's runner fires no exit event, so seeds of killed runs are swept by pid. test/install-fixture.test.ts is the ENG-1 tripwire: put/setVersion/chmod in one clone leave the seed and a sibling unchanged, and clones share no inodes with the read-only seed. * test(free): split claude-overlay-setup by describe (W6b) The 115 s serial file becomes default/lifecycle/installs files (5/4/3 tests, unchanged bodies) sharing test/helpers/claude-overlay-fixture.ts, so the shard packer can balance them. * test(cso): convert cso-contracts.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: cso E2E/judge cases (skill-e2e-cso*, cso-eval). Sentence pins: 16 -> 8. * test(cso): convert cso-preserved.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: cso E2E/judge cases (skill-e2e-cso*, cso-eval). Sentence pins: 10 -> 3. * test(cso): convert cso-eval.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: cso E2E/judge cases (skill-e2e-cso*, cso-eval). Sentence pins: 7 -> 4. * test(browser): convert aside-driver.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: Aside/browse E2E cases; the installer ban stays a meaning check. Sentence pins: 27 -> 9. * test(browser): convert third-party-actions.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: Aside/browse E2E cases; the installer ban stays a meaning check. Sentence pins: 17 -> 2. * test(skills): convert skill-validation.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: workflow judges in skill-llm-eval. Sentence pins: 19 -> 6. * test(skills): convert workflow-excerpt.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: workflow judges in skill-llm-eval. Sentence pins: 37 -> 10. * test(skills): convert question-tuning-registry-path.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: workflow judges in skill-llm-eval. Sentence pins: 4 -> 0. * test(skills): convert host-conformance.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: workflow judges in skill-llm-eval. Sentence pins: 4 -> 4. * test(memory): convert memory-ingest-landing.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: memory-ingest free tests that run the binary. Sentence pins: 11 -> 9. * test(windows): exclude the skip-only claude-overlay setup files from the Windows subset Their real-setup cases are describe.skipIf(win32); after the split their bin/ spawns live in the shared helper, so the source scan stopped seeing them. * test(skills): convert agents-digest.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: the owning skills' E2E cases and workflow judges; CLI/hook output and fixture-data pins in the tail were left alone. Sentence pins: 1 -> 0. * test(skills): convert brain-preflight.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: the owning skills' E2E cases and workflow judges; CLI/hook output and fixture-data pins in the tail were left alone. Sentence pins: 2 -> 1. * test(skills): convert codex-skill-isolation.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: the owning skills' E2E cases and workflow judges; CLI/hook output and fixture-data pins in the tail were left alone. Sentence pins: 1 -> 0. * test(skills): convert office-hours-review.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: the owning skills' E2E cases and workflow judges; CLI/hook output and fixture-data pins in the tail were left alone. Sentence pins: 2 -> 1. * test(skills): convert onboarding-moved-literals.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: the owning skills' E2E cases and workflow judges; CLI/hook output and fixture-data pins in the tail were left alone. Sentence pins: 2 -> 2. * test(skills): convert open-gstack-browser-liveness.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: the owning skills' E2E cases and workflow judges; CLI/hook output and fixture-data pins in the tail were left alone. Sentence pins: 3 -> 2. * test(skills): convert plan-floor-review.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: the owning skills' E2E cases and workflow judges; CLI/hook output and fixture-data pins in the tail were left alone. Sentence pins: 3 -> 2. * test(skills): convert plan-review-report-recording.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: the owning skills' E2E cases and workflow judges; CLI/hook output and fixture-data pins in the tail were left alone. Sentence pins: 4 -> 2. * test(skills): convert regression-1539-review-self-verify.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: the owning skills' E2E cases and workflow judges; CLI/hook output and fixture-data pins in the tail were left alone. Sentence pins: 4 -> 0. * test(skills): convert skill-fixture.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: the owning skills' E2E cases and workflow judges; CLI/hook output and fixture-data pins in the tail were left alone. Sentence pins: 6 -> 3. * test(skills): convert template-context-parity.test.ts sentence pins Safety and authority sentences become expectMentions keyword checks (reword freely, keep the rule); other English sentence pins are dropped. Headings, labels, placeholders, commands, field names, enum/status tokens and ordering checks stay. Behavior and wording: the owning skills' E2E cases and workflow judges; CLI/hook output and fixture-data pins in the tail were left alone. Sentence pins: 1 -> 0. * test: restore pins on renderer output and paid-fixture inputs cso-contracts (the report renderer's required empty phrase), memory-ingest (binary output), carve-plan-fixture, ship-skip-actor and cso-eval (inputs the paid harness builds) assert product or fixture output, not skill prose, so the value-bar narrowing does not apply to them. * feat(eval-policy): v2 series identity, HARNESS_VERSION guard and identity backtest W4a (CEO-27/ENG-3/DX-6), not yet wired to the writer: caseSeriesIdentitiesV2 keys a case by the bytes it owns (paid test file, fixtures, its skills' prompt files including the generated SKILL.md) plus HARNESS_VERSION, and records the full consumed-input fingerprint as provenance. scripts/harness-version.json pins the blob of every shared-harness file (GLOBAL_TOUCHFILES plus the session runners, PTY harness, e2e gate and LLM judge), one line per file; test/harness-version.test.ts fails on drift with both fixes, and scripts/bump-harness-version.ts records the decision: --bump (new series for every case) or --non-behavioral (compatibility entry, no reset). eval-trial-series --backtest replays identities over weekly main revisions and recorded trials (ENG-3/DX-12). * feat(pass-rates): weekly history is main's scheduled runs plus main dispatches W4b: pass-rates reads evals-periodic history from main only, keeping scheduled runs and main dispatches and never a branch dispatch, so back-to-back branch validation runs neither feed pass rates nor count as weeks for quarantine expiry or drift. --branch <name> still reads one branch's runs for inspection. * docs(test-audit): EVAL_POLICY v2 decision memo with the identity backtest D1 memo (CEO-33, ENG-3, DX-12, CEO-27): P(green) arithmetic, false-blocking targets, the v2 identity rule and harness guard, weekly history, W4c evidence (the named 'stochastic' cases pass 26/27 to 34/34, so they stay rule), the backtest over 29 weekly main revisions and 14,898 recorded trials, what it means for quarantine reachability, alternatives (gating vs trended, adaptive panels not adopted, buy vs build), and the rollback proof on the pre-PR reader. * feat(eval-policy): EVAL_POLICY v2 writer: case-owned series identity The v2 writer, last commit of the W4 block (DX-12/ENG-14/ENG-21): the report job's stamp now writes caseSeriesIdentitiesV2 (case-owned bytes plus HARNESS_VERSION) as series_identity, the full consumed-input fingerprint as series_fingerprint and harness_version as provenance, and EVAL_POLICY.version is 2 (approved 2026-10-04, D1). The v1 touchfile identity is removed. Rollback = revert this commit (and the W4 commits above the reader-compat floor); readers then ignore the v2 records with a printed count. * test: restore pins on paid-harness prompt inputs review-enum-lifecycle and plan-review-report-recording assert the prompt a paid case's harness builds (opts.prompt / call.prompt), a fixture contract, not skill prose. * test(security): check injected safety lines on meaning, not capitals content-security, the canary instruction and the arm judge prompt still must carry their never-follow / never-output / exact-"none" rules; the checks are case-insensitive keyword checks so emphasis can change. The UNTRUSTED envelope markers and "none" value stay exact tokens. (Hidden element stripping fails locally only because Playwright Chromium is not installed; same at base.) * test(plan-ceo): keep the outside-voice scope-summary rule as a meaning check The first codemod pass left an empty if body (TS1313). * chore(windows): re-record the Windows duration seed for the 631-file subset (run 37222668939) * ci(free-tests): summarize every shard's flake ledger in the aggregate job (W7a) The aggregate downloads flake-ledger-* (one subdirectory per shard) and appends lane C's `scripts/test-health-report.ts flake-summary` output to the job summary before strict verification. The runner's own flaky table is dropped so the ledger reader stays the single source; --ci-verify keeps the shard-overrun warning. * feat(evals): free test files anywhere and the stored-zip helper never restore the full PR gate (W1a) The CEO-23 replay found design/test free tests forcing fallback (36485436963). * feat(evals): replay recorded PR pushes through today's selector (CEO-23, ENG-18) scripts/replay-pr-selection.ts reports per push the recorded vs replayed mode and case count, dropped gate cases, retired failed cases and every miss (a failed case that depends on the diff but is no longer selected). * refactor(evals): one case-sharded carve-section-loading file with a case id and touchfiles per skill (W5a) * test(free): hold qa-deadline's full-pipe case on a FIFO, not a paused Bun stream Bun gives a child 'pipe' stdio a unix socketpair, so whether the 2 MB fill actually blocks depends on the parent's reader and the host's socket buffer sizes. CI (run 36761671811, both attempts) saw exit 0 after ~130 ms; with net.core.{w,r}mem_default=4 MiB locally the old case fails 3/3 the same way. A FIFO the test never reads is bounded by the kernel pipe buffer, so the receipt write genuinely blocks and the 5 s settlement deadline must fire. The 7 s bound now starts once the preload has queued the write. * test(free): event-driven waits in pty-output-wake close-releases-observers: the 500 ms wall race becomes an ordering check (the waiter is released while only SIGINT has been sent, before close's escalation finishes), and the Bun.spawn spy fakes only the PTY launch so a shard neighbor's in-process spawn cannot add a stray SIGTERM (CI runs 36015568981, 36036582721, 36074636848). continuous-redraws: the fixture spinner stops once the observer has made five coalesced reads, so the run reaches its work deadline with no redraw in flight. Streaming right up to the deadline raced the helper's screen drain and failed as 'PTY screen parse failed; viewport is incomplete' (5 CI runs 09-30..10-03; 4/40 locally under 8 busy loops, 0/40 after). * test(free): session-runner long-capture handshake waits for the collected Read event The fake CLI's init line arms the runner's work-phase timer, so the worker's handshake (work-ready + a >200 s timer) could pass before the runner's read loop had collected the Read event printed just before work-ready. Advancing the clock then killed the run with the Read lost: readSection false (CI run 37194511124). The worker now also waits for the runner's own progress line for that Read. Forcing the observed ordering (work-ready 200 ms before the Read line) fails the old test deterministically and passes the new one. * test(free): budget-override restores the CI variables it rewrites Its provenance cases set and then deleted CI, GITHUB_ACTIONS, GITHUB_REF_NAME, GITHUB_SHA and the GitLab CI_* names without restoring them, so in a CI shard every later file ran with CI unset. qa-only-cleanup's real browse daemons then launched Chromium without --no-sandbox, the sandbox failed on the runner, and both real-Chromium cases timed out with 'QA cleanup fixture did not become ready' (shard 12 attempt 1 of six runs on 10-04, retry alone green). Retained startup log: 'Chromium sandboxing failed!', child env with no CI. Reproduced 3/3 with CI=true, userns denied and budget-override ahead of qa-only-cleanup in one process; 0/3 after. * test(free): pr-shared-input-selection names its base and package history Every selection now passes EVALS_BASE and packageVersionOnly explicitly, so no case reads the checkout's git state. Before, 12 calls ran detectBaseBranch and the cumulative case diffed package.json against merge-base(origin/main, HEAD) — the ambient input that turned main red in runs 36572856858 and 36594314027. A PATH git spy counts 14 git calls in the old file and 0 now. * feat(evals): derive PR-lane dependents for unmapped files instead of restoring the full gate scripts/pr-dependencies.ts derives, from the checked-out tree, which paid cases consume each repository file: test import closures, repo paths and path.join segments in code/shell/skill text (whole-line comments and selection data excluded), bin script names, compiled binaries by outfile path (scripts/build.sh), package scripts, template placeholders through RESOLVERS, and host configs by host id. The global touchfiles' import closure is global. PR selection: an unmapped file with derived consumers puts the run in the new 'dependents' mode (every dependent gate case and judge, no profile cut); a tracked file under a derivable directory that nothing reaches is exempt; periodic/marathon workflows, .gitignore and tsconfig.test.json never run the PR gate. PR_FULL_GATE_FILES is the fail-closed list (deps, CI image, setup, build, evals.yml and its actions, test preload, tsconfig.json, .gitattributes, the skill generator and preamble resolvers, the eval policy). Untracked or underivable files restore the full gate with fix text. The summary and PR comment list derived dependents. * feat(evals): replay misses count derived dependents; report wave vs ordinary pushes (CEO-23) * test: drop FREE_FIXTURES entries for fixtures deleted in W8b * test: re-shape the census attribution fixture to qa-workflow and carve case shards * test: census fixture case order * ci: 420 s PR-lane slices with per-slice ceilings; periodic tracking issue only on main evals.yml joins the census lanes at the 420 s slice budget (W5c/CEO-15) and indexes each eval slice's job timeout from .plan.sliceCiTimeoutMinutes with the slice-deadline clock as its first step (W2c/ENG-2). evals-periodic.yml's tracking-issue upsert and close run only on refs/heads/main, so a branch dispatch whose gate census is skipped (CEO-04) never opens or closes #2755. * docs: integrate lane docs for the audit wave (eval-bg backends, PR fallback, policy v2, ceilings, evidence) - CLAUDE.md: 'Running evals as an agent' describes scripts/eval-bg.ts's dispatch and local backends, forcing flags, the sentinel and status; eval:select's PR-profile default; prompt-structure helper and value-bar terms. instruction-facts now requires scripts/eval-bg.ts in that section. - CONTRIBUTING.md: decision table gains test:health, eval:select default, eval:bg backends and the include_gate_census / evals_all dispatch lines; 420 s runners and per-runner ceilings; flake ledger, OSV tracking issue and expiry test, weekly test health; prompt-structure guidance. - AGENTS.md: build commands name the eval:bg backend choice and test:health. - docs/TESTING_INTERNALS.md: anchors the fix texts point to (#pr-paid-lane-fallback, #pass-rate-policy-versions, #harness-version, #flake-ledger, #duration-seed), receipts/recovery/debounce, per-slice ceilings, the slice deadline with not_run/hung and checkpoints, the eval-bg cap formula in place of the detach floors, EVAL_POLICY v2 history, carve case shards via --case. - docs/test-audit-2026-10.md: CEO-23 replay (165/168 -> 164/168 full-fallback, 0 misses), ENG-17 slice table (re-measured on this tree), D2 prose-pin metric. - TODOS.md: progress note on 'Let pass-rate history decide the rest'. - test/testing-internals-anchors.test.ts: every TESTING_INTERNALS.md#anchor named in code must match a heading. * test(free): hold qa-deadline's blocked-output-pipe receipts on a FIFO The five 'guard %s receipts survive a blocked output pipe' cases filled a paused Bun child pipe with 2 MiB. That pipe is a unix socketpair, so the fill only blocks while the host's socket buffers are smaller than the write: with net.core.{w,r}mem_default=4 MiB the old cases fail 13 of 15 case runs (the guard finishes before the test resumes the stream). The held stream is now a FIFO the test drains only after resume(), bounded by the kernel pipe buffer; Windows keeps the paused pipe. Bounds and assertions are unchanged. * test(free): qa-deadline waits for a written pid, not just the pid file DRIVER and LEAF write their pid with writeFileSync, so under load ready() can see the file before its digits and fail 'expected > 0, received 0' (seen once in 150 loaded runs). After any early failure the finally SIGKILLs the CLI, which orphans the detached command group; a leaf whose pid file is not yet readable then keeps the output stream open and the case hangs to its 15 s timeout (2/80 old-file runs and 3/310 FIFO-only runs under 8 busy loops; the instrumented hang showed the CLI already SIGKILLed by the finally). ready() now polls within the same 3 s bound until the file holds a positive pid. * fix(evals): a deleted file selects the dependents of the live files that still reference it PR selection treated every path absent from the head tree as an unknown dependency, so any PR that deleted a file forced the full gate. A deleted path is now placed like a live one, against the head tree: a matching touchfile selects its cases; live code, skill text or a full-gate input that still references it (path, extensionless import target, relative import, bin name, ancestor directory, compiled-binary outfile; whole-line comments and data files excluded) contributes its dependents, a full-gate referencer restores the full gate with a 'deleted but still referenced' fix, and an unreferenced deleted path has no dependents. Deleted PR_FULL_GATE_FILES entries still force the full gate. Tests that used absent paths as stand-ins for new unmapped files now use tracked root files (or an explicit tracked set); the CEO-23 replay counts misses through those references too. * docs: integrate free-suite lane docs (seed anchor, Windows lane, native qualification, install fixture, flake rules) - docs/TESTING_INTERNALS.md: '### Free suite duration seed' (the anchor the seed messages link) with DX-13, the 99th-percentile packing, --ci-plan warning, --ci-verify overruns and the CEO-10 seed ratchet; the Windows 6-job lane, its Windows-measured seed and refresh dispatch, curation and WINDOWS_CURATED_FLOOR; native-qualification.yml modes; the merged install fixture; test-only timing knobs; the flake-ledger job summary, Windows ledger names and three free-suite flake rules; zsh in the Ubicloud mirror. - scripts/test-health-report.ts links #free-suite-duration-seed, so the two seed messages share one heading (the anchor test caught the split). - CONTRIBUTING.md: Windows seed refresh and native qualification rows; seed-source pointer. AGENTS.md: Windows six-job lane. - docs/cso-release-qualification.md: the free CSO Docker job runs only cso-docker-integration.test.ts. - docs/test-audit-2026-10.md: free-suite evidence lines. - TODOS.md: CEO-11b progress note on the pre-push hook hang. * test: final duration seeds on the integrated tree; retire the emptied cookie-validation-phases file Linux seed re-recorded on Ubicloud standard-16 and the Windows seed from dispatch run 37227140871, both atc903310. The seed allowlist drops the three entries that are no longer over the ceiling. cookie-validation-phases lost its validation-phase pins with W8f and its Windows/Dia pins moved to windows-native-workflows with native-qualification.yml, so the file is deleted and the curated Windows floor rises to 638. The Windows recorder writes the seed at its default path so GSTACK_FREE_TEST_DURATIONS never leaks into timed tests. * test: re-express stale PR-lane pins against the new evals.yml contract - ci-eval-cache: the profile/fresh expressions now include evals_all=false dispatches and pr_receipts/evals-fresh; the PR store is restored only by recover-receipts (still scoped to repository_id + PR, PR events only), plan-slices takes it as an artifact and never touches the cache, saves stay PR-only, EVALS_CACHE_PR binds pr_receipts dispatches, and the report verdict artifact carries report receipts. - ci-paid-coordination: plan-slices needs only the two ubuntu coordination jobs (never the image); the fork/Dependabot clause is unchanged on plan-slices and also gates debounce and recover-receipts; only recovery holds actions: read. - qa-supervision-selection: auq-consistency and auq-verbose-vs-carved-ab (W1b) reach claude-pty-runner and pty-screen through auq-sdk-capture, exactly as auq-format-gate does. - paid-retry-supervision: the full-validation plan uses a tracked full-gate input (setup); an absent path is now a deletion placed by its references. * ci(windows): record the Windows seed at the default path * chore: HARNESS_VERSION non-behavioral decision for the audit-wave integration * v1.91.20.0 release metadata: test, eval and CI audit wave * test(compare-board): wait for the post-submit state the feedback POST settles The two post-submit assertions read the board once, but the board disables the submit button and shows the message only after its feedback POST resolves. On a loaded Ubicloud shard the read beat the round trip (FLAKY-PASS in the final free-suite run). Delaying the POST by 400 ms reproduces both failures at HEAD; polling for the settled state passes. * test(plan-eng plan-mode): run the seeded STOP-gate smoke in its own fixture project The seeded plan proposes a custom test runner, so the review's Step 0 reads the working tree's test setup. In the gstack checkout that is gstack's own sharded runner, and the case timed out at 300 s in census 37228573062 after this wave reshaped the runner. It now runs in a four-file Bun project with git history, with the product-notice suppression auto-decide-preserved's standalone fixture already uses. Local pass on Bun 1.4.0 + Claude Code 2.1.284: asked in 121 s, seed tokens observed. * test(plan-mode evidence): pin the seeded eng smoke's isolated fixture cwd and env * ci(image): Git >= 2.44 so paid shared-libs cases run the supported safe-git path bin/gstack-safe-git prefixes every /deslop-shared-libs Git read with --no-lazy-fetch, which noble's git 2.43 rejects (exit 129). The eval image therefore ran every paid shared-libs case on the unsupported-Git fallback: local history unavailable, branch HEAD unknown. shared-libs-read-only failed on exactly that in PR run 37230639146 ("Current branch HEAD: unknown. The safe-git probe failed (unknown option: --no-lazy-fetch, git 2.43.0)") and in 2 of 74 earlier PR runs. The image installs current Git from the Ubuntu Git Maintainers PPA and fails its build if --no-lazy-fetch is missing; the deliberate unsupported-Git case keeps its fake Git. * eval: pool matching branch census trials into main's pass-rate series (D1 option b) Approved 2026-10-05. eval:pass-rates also reads completed branch census runs in the weekly window. A branch trial counts only toward a series a main run also has (same case-owned bytes, HARNESS_VERSION, model and CLI), so a branch extends main's history but never starts or becomes a case's current series; weeks for quarantine expiry still count main runs only. The D1 backtest put this at 67 of 210 cases with enough trials for the entry rule, against 0 of 195 on main alone. * C2: /qa defines its outcome filename fields; QA report templates name the measured timing fields Low judge samples in failing panels named both: qa workflow (37186854666) 'an undefined {user} in the outcome filename', and qa-only workflow (37198445662) 'template asks for Duration / durationMs totals while reporting.md mandates Probe budget and Guarded command time'. The qa union stays under its 1.11 size ratio (93292 of 93294 bytes). (cherry picked from commite14c0c1de1) * C2: Browser Quick gets a feasible budget; QA loop names its directory and stop path Judge rationales on the qa-only panels (2 of the last 3) cited the 30 s Browser Quick budget: homepage + 5 targets, each a checkpoint write and a guarded command, left ~5 s per probe. Quick is now 180 s (method text: 3 minutes); the qa-only-no-fix paid check expects the same native deadline. PROBE_DIR points at its source line, a stopped loop goes to the final report steps (annotations, materialize, Markdown), and /review and /ship say that after smoke expiry only plan checks rerun in revalidation. (cherry picked from commit78d4a2a343) * qa: fit the C2 judge fixes under the qa and review size caps Brings GSTA-21's qa.ts trims (3bd625d,ce73af0) for the cherry-picked C2 wording: PROBE_DIR keeps 'owned', the stop line points at §4, and the post-expiry rule is one sentence, so /review fits its cap. qa measures 93,303 / 84,049 = 1.11010 against the same 1.1102 cap GSTA-21 set. The qa-only workflow judge passed 10 of 10 local trials on this wording (7 of 10 before the C2 fixes). * test: qa-only fixture keeps its caller-set 30 s deadline The qa-only-no-fix fixture tells the model to start the guard with 30 s (a caller limit), so its deadline assertion's default stays 30 000 ms; the product's Browser Quick default (180 s) does not apply to it. Reverts the helper and its test to their base. (cherry picked from commite36e622add) * test: qa-quick sets a caller probe limit and drains past its session budget C2 raised Browser Quick's native budget to 180 s. qa-quick is the only gate case on the native budget, and its wall went 194 s (main periodic, pre-C2) -> 254 s and 274 s (this PR's runs) -> 300 s session timeout (run 37242691913): the agent now probes every navigation target with a checkpoint, then needs ~80 s to report. Like qa-only-no-fix, the eval now sets a caller probe limit (30 s), scopes the fixture, and gives bun's outer timeout the session drain margin instead of racing the 300 s session. The product's 180 s budget and the session budget are unchanged. Diagnostic rerun passed in 195 s. (cherry picked from commitf8dc64031e) * chore: restore the audit wave's package scripts after the main merge (version 1.91.23) * qa: keep main's one-sentence post-expiry rule for /review and /ship The C2 variants ("step 4c reruns plan checks only" and "even in 4c") read to the review workflow judge as conflicting with Step 5.8's repeat-pass reruns: 6 of 10 local trials passed with the longer sentence and PR run 37248650879 failed on the shorter one. Main's "Post-expiry smoke rechecks are not-run." passed 10 of 10. qa (9/10), qa-only (10/10) and ship-managed-hook-refresh (10/10) were measured on the same tree. * v1.91.24.0: move off the slot #3036 claims
862 lines
52 KiB
TypeScript
862 lines
52 KiB
TypeScript
/**
|
|
* Paid-lane planner: duration seeds, slice packing, the run manifest and its verification. Moved from scripts/test-paid-shards.ts.
|
|
*/
|
|
import * as fs from 'node:fs';
|
|
import * as path from 'node:path';
|
|
import { normalizeRelativePath } from './shard-engine';
|
|
import { CASE_QUARANTINE } from '../../test/helpers/periodic-exclude-data';
|
|
import { FILE_RETRY_BUDGETS, STRICT_RETRY_CASE_BUDGETS } from '../../test/helpers/eval-budgets';
|
|
import { evalEntryOutcome, type EvalCaseKind } from '../../test/helpers/eval-store';
|
|
import { E2E_KINDS } from '../../test/helpers/touchfiles-data';
|
|
import { prProfileCaseAllowed, prProfileFileMap, type PrProfileSelection } from '../test-pr-profile';
|
|
import { E2E_TOUCHFILES, E2E_TIERS, LLM_JUDGE_TOUCHFILES } from '../../test/helpers/touchfiles';
|
|
import { CASE_KEY_SEPARATOR, CASE_SHARDED_FILES, type CaseTrialPlan, caseTrialPlan, codexShardAccess, expandCaseShards, expandTrialShards, isIsolatedCase, partitionCaseExclusions, sameTrialPlan, shardCaseId, shardFile, shardTrial } from './paid-cases';
|
|
import { DEFAULT_JOBS, OVERLAY_MAX_ACTIVE_SHARDS, PAID_TIERS, SLICE_UPLOAD_RESERVE_MS, type PaidCaseSelection, type PaidProfile, type PaidShardBudget, type PaidTier, ROOT, type ShardOutcome, type ShardStatus, type ShardTrialRecord, isOverlayTestFile } from './paid-types';
|
|
import { collectPaidTestFiles, computePaidCaseSelection, expectedPrCaseCount, isAllSkippedPass, paidShardWallUpperBoundMs, partitionShardsByDiffSelection, planPaidShards, prProfileFileSelected, resolvePaidShardBudget, resolvePaidShardTimeoutMs, sameBudget, selectPaidTestFiles, shardSlug, validatedProfile } from './paid-select';
|
|
|
|
// ─── Planner / executor / report (the CI re-platform surface) ──────────────
|
|
// One PLANNER computes selection and the slice plan ONCE; K executor jobs
|
|
// consume it; a REPORT reconciles results against the plan. This kills two
|
|
// classes at the root: per-slice selector divergence (one slice failing
|
|
// merge-base resolution and running a different partition than its siblings)
|
|
// and hollow lanes (a missing/failed slice that artifact-presence aggregation
|
|
// would read as green). CI wiring: evals.yml planner job → K-way matrix of
|
|
// `--plan manifest.json --slice i` → report job running `--report <dir>`.
|
|
|
|
export interface ManifestEntry {
|
|
file: string;
|
|
/** 1-based executor slice for planned entries; 0 for skipped/excluded. */
|
|
slice: number;
|
|
status: 'planned' | 'skipped-by-diff' | 'excluded';
|
|
reason?: string;
|
|
/** Required when a registered retry-budget file is planned. */
|
|
budget?: PaidShardBudget;
|
|
/** Budget-mode packing weight (recorded wall, or the whole budget when unknown). */
|
|
estimatedMs?: number;
|
|
/** Isolated trial shard (`<file>#<id>~t<N>`): the case's kind, fixed panel and quarantine at plan time. */
|
|
trial?: CaseTrialPlan;
|
|
/** File shard whose isolated cases run as trial shards: their ids, excluded by name here. */
|
|
excludeCases?: string[];
|
|
}
|
|
|
|
/** Budget-mode plan: per-executor estimate and the CI job timeout it needs. */
|
|
export interface PaidSlicePlan {
|
|
sliceBudgetMs: number;
|
|
jobs: number;
|
|
estimatedSliceMs: number[];
|
|
/** The largest per-slice ceiling: one number for executors that use a single job timeout. */
|
|
ciTimeoutMinutes: number;
|
|
/** Per-slice CI job ceilings (sliceCiTimeoutMinutes); absent in plans written before W2c. */
|
|
sliceCiTimeoutMinutes?: number[];
|
|
/** Slices that run only Codex shards; CI executes them on the host-run Codex job, where bubblewrap can start. */
|
|
codexSlices?: number[];
|
|
}
|
|
|
|
export interface PaidRunManifest {
|
|
version: 1;
|
|
tier: PaidTier;
|
|
evalsAll: boolean;
|
|
sliceCount: number;
|
|
selectionReason: string;
|
|
/** Legacy v1 manifests omit these; new plans bind case-level execution. */
|
|
profile?: PaidProfile;
|
|
selection?: PaidCaseSelection;
|
|
prCoverage?: PrProfileSelection;
|
|
plan?: PaidSlicePlan;
|
|
entries: ManifestEntry[];
|
|
}
|
|
|
|
/**
|
|
* Paid evals never retry (approved 2026-09-29): a failed verdict is final for
|
|
* its run, and trials are fixed by kind before the run (EVAL_POLICY). The
|
|
* function stays the single statement of that policy for the Bun arguments
|
|
* and the reuse identity.
|
|
*/
|
|
export function retriesForFiles(_files: string[]): number {
|
|
return 0;
|
|
}
|
|
|
|
export const PAID_TEST_DURATIONS_FILE = 'scripts/paid-test-durations.json';
|
|
|
|
/**
|
|
* Recorded per-file paid-shard wall times (ms) per tier from real CI slice
|
|
* reports (a file's gate and periodic cases differ), refreshed with
|
|
* `--report <dir> --write-durations`. A packing hint only: a missing or
|
|
* corrupt seed keeps the supervision-budget allocation.
|
|
*/
|
|
export function loadPaidTestDurations(rootDir = ROOT, tier: PaidTier = 'gate'): Record<string, number> {
|
|
try {
|
|
const parsed = JSON.parse(fs.readFileSync(path.join(rootDir, PAID_TEST_DURATIONS_FILE), 'utf8')) as { version?: unknown; tiers?: Record<string, Record<string, unknown>> };
|
|
if (parsed.version !== 2) return {};
|
|
return Object.fromEntries(Object.entries(parsed.tiers?.[tier] ?? {})
|
|
.filter((entry): entry is [string, number] => typeof entry[1] === 'number' && Number.isFinite(entry[1]) && entry[1] > 0));
|
|
} catch {
|
|
return {};
|
|
}
|
|
}
|
|
|
|
/** Rewrite one tier of the committed seed, keeping the other tiers. */
|
|
export function writePaidTestDurations(tier: PaidTier, durations: Record<string, number>, rootDir = ROOT): void {
|
|
const target = path.join(rootDir, PAID_TEST_DURATIONS_FILE);
|
|
const tiers = Object.fromEntries(PAID_TIERS.map(name => [name, loadPaidTestDurations(rootDir, name)])
|
|
.filter(([name, recorded]) => name === tier || Object.keys(recorded as object).length > 0));
|
|
tiers[tier] = durations;
|
|
const temporary = `${target}.tmp-${process.pid}`;
|
|
fs.writeFileSync(temporary, `${JSON.stringify({ version: 2, recordedAt: new Date().toISOString(), tiers }, null, 2)}\n`);
|
|
fs.renameSync(temporary, target);
|
|
}
|
|
|
|
/** Seed key of a shard: trials of one case share their case key (`<file>#<id>`). */
|
|
function durationKey(key: string): string {
|
|
const rel = normalizeRelativePath(key);
|
|
return shardTrial(rel) === null ? rel : `${shardFile(rel)}${CASE_KEY_SEPARATOR}${shardCaseId(rel)}`;
|
|
}
|
|
|
|
/** Recorded wall of a shard key; an unrecorded trial falls back to its whole file's wall. */
|
|
export function recordedShardMs(recorded: Record<string, number>, key: string): number | undefined {
|
|
const rel = normalizeRelativePath(key);
|
|
return recorded[durationKey(rel)] ?? (shardTrial(rel) === null ? undefined : recorded[shardFile(rel)]);
|
|
}
|
|
|
|
/** Longest recorded wall a PR-lane (gate) shard may have: the lane's ~10-minute case target. */
|
|
export const PR_LANE_SHARD_LIMIT_MS = 600_000;
|
|
|
|
/** One line per planned shard whose recorded wall exceeds `limitMs` (split or shorten it; never raise the limit). */
|
|
export function shardDurationViolations(keys: readonly string[], recorded: Record<string, number>, limitMs = PR_LANE_SHARD_LIMIT_MS): string[] {
|
|
return keys.flatMap(key => {
|
|
const ms = recorded[durationKey(key)];
|
|
return ms !== undefined && ms > limitMs ? [`${key}: recorded ${Math.round(ms / 1000)}s > ${limitMs / 1000}s`] : [];
|
|
});
|
|
}
|
|
|
|
/**
|
|
* Merge a report's executed single-file outcomes into the seed; all-skipped
|
|
* shards carry no cost signal. Trials of one case record their longest wall
|
|
* under the case key.
|
|
*/
|
|
export function mergePaidTestDurations(seed: Record<string, number>, results: SliceResult[]): Record<string, number> {
|
|
const merged = { ...seed };
|
|
const fresh = new Map<string, number>();
|
|
for (const result of results) {
|
|
for (const outcome of result.outcomes) {
|
|
if (outcome.files.length !== 1 || outcome.elapsedMs < 1_000 || isAllSkippedPass(outcome) || outcome.reused) continue;
|
|
const key = durationKey(outcome.files[0]!);
|
|
fresh.set(key, Math.max(fresh.get(key) ?? 0, outcome.elapsedMs));
|
|
}
|
|
}
|
|
for (const [key, ms] of fresh) merged[key] = ms;
|
|
return Object.fromEntries(Object.entries(merged).sort(([a], [b]) => (a < b ? -1 : 1)));
|
|
}
|
|
|
|
/** True when a shard key's file sees the CI Codex install in `tier` (scripts/lib/paid-cases.ts codexShardAccess). */
|
|
export function isCodexShard(key: string, tier: PaidTier): boolean {
|
|
return codexShardAccess([shardFile(key)], tier) !== 'none';
|
|
}
|
|
|
|
/** Sorted slices holding at least one planned Codex shard. */
|
|
export function codexSlicesOf(tier: PaidTier, entries: readonly ManifestEntry[]): number[] {
|
|
return [...new Set(entries.filter(entry => entry.status === 'planned' && isCodexShard(entry.file, tier)).map(entry => entry.slice))].sort((a, b) => a - b);
|
|
}
|
|
|
|
/** Panel identity of a trial shard key (`<file>#<id>`), else null. */
|
|
export function trialPanelKey(key: string): string | null {
|
|
return shardTrial(key) === null ? null : durationKey(key);
|
|
}
|
|
|
|
/** True when `key` is a trial whose panel already has a trial in `planned`. */
|
|
function sharesPanel(planned: readonly string[], key: string): boolean {
|
|
const panel = trialPanelKey(key);
|
|
return panel !== null && planned.some(other => other !== key && trialPanelKey(other) === panel);
|
|
}
|
|
|
|
/** Setup, image pull and artifact upload allowance on top of a slice's supervised wall. */
|
|
export const CI_SETUP_ALLOWANCE_MINUTES = 20;
|
|
|
|
/**
|
|
* CI job ceiling of one slice (W2c, ENG-2): its envelope is the larger of twice
|
|
* the slice budget, the longest single shard's supervised wall, and, for the
|
|
* overlay slice, the serialized overlay group; plus the setup allowance. A hang
|
|
* fails within this bound instead of holding a runner for the sum of every
|
|
* shard's worst case; the executor's in-process deadline (sliceDeadlineMs)
|
|
* stops starting work 5 minutes earlier so results still upload. Same cases run.
|
|
*/
|
|
export function sliceCiTimeoutMinutes(files: string[], budgetMs: number, jobs: number, timeoutMs?: number): number {
|
|
const longestShardMs = Math.max(0, ...files.map(file => resolvePaidShardTimeoutMs([file], timeoutMs)));
|
|
const overlayEnvelopeMs = files.some(isOverlayTestFile) ? sliceSupervisedWallMs(files, jobs, timeoutMs) : 0;
|
|
return Math.ceil(Math.max(2 * budgetMs, longestShardMs, overlayEnvelopeMs) / 60_000) + CI_SETUP_ALLOWANCE_MINUTES;
|
|
}
|
|
|
|
/**
|
|
* The executor's in-process deadline: job start (the workflow's first step
|
|
* records it in GSTACK_SLICE_JOB_STARTED_AT, epoch seconds) plus the slice's
|
|
* job ceiling, minus the upload reserve. Without a recorded job start (local
|
|
* runs) the executor's own start stands in.
|
|
*/
|
|
export function sliceDeadlineMs(ceilingMinutes: number, jobStartedAtSeconds: string | undefined, fallbackStartMs: number): number {
|
|
const recorded = Number(jobStartedAtSeconds);
|
|
const start = Number.isSafeInteger(recorded) && recorded > 0 ? recorded * 1000 : fallbackStartMs;
|
|
return start + ceilingMinutes * 60_000 - SLICE_UPLOAD_RESERVE_MS;
|
|
}
|
|
|
|
/** Estimated wall of one executor running `files` in order on `jobs` FIFO workers. */
|
|
export function estimatedSliceMs(files: string[], weight: (file: string) => number, jobs: number): number {
|
|
const workers = Array<number>(Math.max(1, jobs)).fill(0);
|
|
for (const file of files) {
|
|
const next = workers.indexOf(Math.min(...workers));
|
|
workers[next] += weight(file);
|
|
}
|
|
return Math.max(...workers);
|
|
}
|
|
|
|
/** Executor order within a slice: longest recorded work first, then by path. */
|
|
export function sliceExecutionOrder<T extends { file: string; estimatedMs?: number }>(entries: T[]): T[] {
|
|
return [...entries].sort((a, b) => (b.estimatedMs ?? 0) - (a.estimatedMs ?? 0) || (a.file < b.file ? -1 : a.file > b.file ? 1 : 0));
|
|
}
|
|
|
|
/** Supervised worst case of one slice in execution order, overlays at their own admission limit. */
|
|
export function sliceSupervisedWallMs(files: string[], jobs: number, overrideMs?: number): number {
|
|
return paidShardWallUpperBoundMs(files.filter(file => !isOverlayTestFile(file)), jobs, overrideMs)
|
|
+ paidShardWallUpperBoundMs(files.filter(isOverlayTestFile), Math.min(jobs, OVERLAY_MAX_ACTIVE_SHARDS), overrideMs);
|
|
}
|
|
|
|
/**
|
|
* Budget packing: one runner per file or per tightly packed group. Files go
|
|
* longest-recorded-first into the fullest slice whose estimated wall stays
|
|
* within the budget (best fit), else into a new slice. A file with no recorded
|
|
* wall weighs the whole budget, so unknown cost gets a runner of its own. A
|
|
* file longer than the budget runs alone. Overlay wrappers keep one shared
|
|
* final slice (one wrapper at a time). Codex shards never share a slice with
|
|
* other shards. Each slice gets its own CI job ceiling (sliceCiTimeoutMinutes).
|
|
*/
|
|
export function packBySliceBudget(files: string[], budgetMs: number, jobs: number,
|
|
recorded: Record<string, number>, timeoutMs?: number, codexShard: (file: string) => boolean = () => false): {
|
|
slices: string[][]; estimates: Record<string, number>; estimatedSliceMs: number[]; ciTimeoutMinutes: number; sliceCiTimeoutMinutes: number[];
|
|
} {
|
|
const estimates = Object.fromEntries(files.map(file => [file, recordedShardMs(recorded, file) ?? budgetMs]));
|
|
const weight = (file: string) => estimates[file]!;
|
|
const slices: string[][] = [];
|
|
for (const file of sliceExecutionOrder(files.filter(file => !isOverlayTestFile(file)).map(file => ({ file, estimatedMs: weight(file) }))).map(entry => entry.file)) {
|
|
let best = -1, bestMs = -1;
|
|
slices.forEach((planned, index) => {
|
|
// Trials of one case never share a runner: independent machines, and
|
|
// the panel's wall stays one trial long.
|
|
if (sharesPanel(planned, file)) return;
|
|
// Codex shards run on the host-run Codex job; other shards keep the default container sandbox.
|
|
if (codexShard(planned[0]!) !== codexShard(file)) return;
|
|
const ms = estimatedSliceMs([...planned, file], weight, jobs);
|
|
if (ms <= budgetMs && ms > bestMs) { best = index; bestMs = ms; }
|
|
});
|
|
if (best < 0) slices.push([file]);
|
|
else slices[best]!.push(file);
|
|
}
|
|
const overlays = files.filter(isOverlayTestFile).sort();
|
|
if (overlays.length) slices.push(sliceExecutionOrder(overlays.map(file => ({ file, estimatedMs: weight(file) }))).map(entry => entry.file));
|
|
if (!slices.length) slices.push([]);
|
|
const estimatedSliceMsList = slices.map(planned => planned.some(isOverlayTestFile)
|
|
? estimatedSliceMs(planned, weight, Math.min(jobs, OVERLAY_MAX_ACTIVE_SHARDS)) : estimatedSliceMs(planned, weight, jobs));
|
|
const ceilings = slices.map(planned => sliceCiTimeoutMinutes(planned, budgetMs, jobs, timeoutMs));
|
|
return { slices, estimates, estimatedSliceMs: estimatedSliceMsList, ciTimeoutMinutes: Math.max(...ceilings), sliceCiTimeoutMinutes: ceilings };
|
|
}
|
|
|
|
/** Worker counts whose worst-case slice wall duration packing may never worsen. */
|
|
export const SUPERVISED_WORKER_COUNTS = [1, 2, 3, 4] as const;
|
|
|
|
/**
|
|
* Allocate the RUNNABLE shard plan across K slices — deterministic. Registered
|
|
* long files are spread by supervision budget and the rest round-robin; that
|
|
* baseline fixes each slice's worst-case wall. With a duration seed, files are
|
|
* then re-packed longest-recorded-first onto the lightest slice, accepting a
|
|
* placement only if no slice's worst-case wall exceeds the baseline's maximum
|
|
* for any supervised worker count. If any file cannot be placed, the baseline
|
|
* stands.
|
|
*/
|
|
export function buildRunManifest(opts: {
|
|
tier: PaidTier;
|
|
profile?: PaidProfile;
|
|
/** Fixed slice count; exclusive with sliceBudgetMs. */
|
|
sliceCount?: number;
|
|
/** Budget mode: pack recorded work so each executor's estimated wall stays
|
|
* within this budget; the slice count follows from the plan. */
|
|
sliceBudgetMs?: number;
|
|
/** Shard workers per executor (EVALS_JOBS) that budget mode plans for. */
|
|
jobs?: number;
|
|
evalsAll: boolean;
|
|
timeoutMs?: number;
|
|
discovered?: string[];
|
|
env?: NodeJS.ProcessEnv;
|
|
rootDir?: string;
|
|
changedFiles?: string[];
|
|
/** Recorded per-file durations; defaults to the committed seed under rootDir. */
|
|
durations?: Record<string, number>;
|
|
/** Weekly gate census only: LLM judges already run in the periodic census and PR gate lanes. */
|
|
skipJudges?: boolean;
|
|
/** Injectable registries (default: E2E_KINDS and CASE_QUARANTINE). */
|
|
kinds?: Record<string, EvalCaseKind>;
|
|
quarantine?: Record<string, unknown>;
|
|
}): PaidRunManifest {
|
|
const budgetMode = opts.sliceBudgetMs !== undefined;
|
|
if (budgetMode === (opts.sliceCount !== undefined)) throw new Error('Plan with exactly one of --slices or --slice-budget');
|
|
if (budgetMode && (!Number.isSafeInteger(opts.sliceBudgetMs) || opts.sliceBudgetMs! <= 0 || !Number.isSafeInteger(opts.jobs) || opts.jobs! <= 0)) {
|
|
throw new Error('--slice-budget needs a positive budget and an explicit positive --jobs');
|
|
}
|
|
if (!budgetMode && (!Number.isInteger(opts.sliceCount) || opts.sliceCount! <= 0)) {
|
|
throw new Error(`--slices needs a positive integer. Received: ${opts.sliceCount}`);
|
|
}
|
|
const rootDir = opts.rootDir ?? ROOT;
|
|
const env = opts.env ?? process.env;
|
|
const profile = opts.profile ?? validatedProfile(env.EVALS_PROFILE, 'EVALS_PROFILE');
|
|
if (profile === 'pr' && opts.tier !== 'gate') throw new Error('PR profile requires gate tier; use --profile full for periodic coverage');
|
|
const discovered = opts.discovered ?? collectPaidTestFiles(rootDir);
|
|
const tierSelection = selectPaidTestFiles(discovered, opts.tier, rootDir);
|
|
const judge = (file: string) => /^test\/skill-llm-eval[^/]*\.test\.ts$/.test(normalizeRelativePath(file));
|
|
const selected = opts.skipJudges ? tierSelection.selected.filter(file => !judge(file)) : tierSelection.selected;
|
|
const kinds = opts.kinds ?? E2E_KINDS;
|
|
const quarantine = opts.quarantine ?? CASE_QUARANTINE;
|
|
const notLive = [...Object.keys(kinds).filter(id => kinds[id] === 'behavior'), ...Object.keys(quarantine)].filter(id => !Object.hasOwn(E2E_TIERS, id));
|
|
if (notLive.length) throw new Error(`Only live E2E cases can be behavior or quarantined (judges sample their panel inside the case): ${notLive.join(', ')}`);
|
|
const caseKeys = partitionCaseExclusions(expandCaseShards(selected, opts.tier, rootDir));
|
|
const excluded = [...tierSelection.excluded, ...(opts.skipJudges ? tierSelection.selected.filter(judge)
|
|
.map(file => ({ file, reason: 'skipped: LLM judges run in the periodic census and PR gate lanes' })) : []), ...caseKeys.excluded];
|
|
const expansion = expandTrialShards(caseKeys.runnable, opts.tier, rootDir, { kinds, quarantine });
|
|
const excludeOf = (key: string) => expansion.excludeCases[normalizeRelativePath(key)] ?? [];
|
|
const shards = planPaidShards(expansion.keys, { maxFilesPerShard: 1 });
|
|
const cases = computePaidCaseSelection({ profile, env, rootDir, changedFiles: opts.changedFiles });
|
|
const fast = cases.coverage?.mode === 'pr';
|
|
const profileShards = fast ? shards.filter(files => prProfileFileSelected(files[0], cases.selection, excludeOf(files[0]!))) : shards;
|
|
const { runnable, skipped } = partitionShardsByDiffSelection(profileShards,
|
|
cases.selection.e2e === null ? null : new Set(cases.selection.e2e), { rootDir, excludeCases: expansion.excludeCases });
|
|
if (fast) for (const files of shards) {
|
|
if (!prProfileFileSelected(files[0], cases.selection, excludeOf(files[0]!))) skipped.push({ files, reason: 'Outside the fast PR profile; retained in broad gate/periodic coverage' });
|
|
}
|
|
const extras = (key: string): Pick<ManifestEntry, 'trial' | 'excludeCases'> => {
|
|
const trial = expansion.trials[normalizeRelativePath(key)];
|
|
const exclude = expansion.excludeCases[normalizeRelativePath(key)];
|
|
return { ...(trial ? { trial } : {}), ...(exclude ? { excludeCases: exclude } : {}) };
|
|
};
|
|
|
|
const entries: ManifestEntry[] = [];
|
|
if (budgetMode) {
|
|
const plan = packBySliceBudget(runnable.map(files => files[0]!), opts.sliceBudgetMs!, opts.jobs!,
|
|
opts.durations ?? loadPaidTestDurations(rootDir, opts.tier), opts.timeoutMs, file => isCodexShard(file, opts.tier));
|
|
plan.slices.forEach((files, index) => files.forEach(file => entries.push({ file, slice: index + 1, status: 'planned',
|
|
estimatedMs: plan.estimates[file]!, ...extras(file),
|
|
...(FILE_RETRY_BUDGETS.some(budget => budget.file === shardFile(file)) ? { budget: resolvePaidShardBudget([file], opts.timeoutMs) } : {}) })));
|
|
for (const s of skipped) entries.push({ file: s.files[0], slice: 0, status: 'skipped-by-diff', reason: s.reason, ...extras(s.files[0]!) });
|
|
for (const e of excluded) entries.push({ file: e.file, slice: 0, status: 'excluded', reason: e.reason });
|
|
entries.sort((a, b) => (a.file < b.file ? -1 : 1));
|
|
return parseRunManifest(JSON.stringify({
|
|
version: 1, tier: opts.tier, evalsAll: opts.evalsAll, sliceCount: plan.slices.length,
|
|
selectionReason: cases.reason, profile, selection: cases.selection,
|
|
...(cases.coverage ? { prCoverage: cases.coverage } : {}),
|
|
plan: { sliceBudgetMs: opts.sliceBudgetMs!, jobs: opts.jobs!, estimatedSliceMs: plan.estimatedSliceMs, ciTimeoutMinutes: plan.ciTimeoutMinutes,
|
|
sliceCiTimeoutMinutes: plan.sliceCiTimeoutMinutes, codexSlices: codexSlicesOf(opts.tier, entries) },
|
|
entries,
|
|
} satisfies PaidRunManifest));
|
|
}
|
|
const sliceCount = opts.sliceCount!;
|
|
const overlaySlice = sliceCount;
|
|
const reserveOverlaySlice = overlaySlice > 1 && runnable.some(files => files.some(isOverlayTestFile));
|
|
const ordinarySlices = overlaySlice - Number(reserveOverlaySlice);
|
|
// Spread registered long files by supervised load. Keep one ordinary-only
|
|
// lane when possible, so every lane does not inherit a long-workflow tail.
|
|
// The reserved overlay slice retains its ownership.
|
|
const ordinary = runnable.filter(files => !files.some(isOverlayTestFile));
|
|
const registered = ordinary.filter(files => FILE_RETRY_BUDGETS.some(budget => budget.file === shardFile(files[0]!)));
|
|
const allocations = new Map<string, number>();
|
|
if (registered.length && ordinarySlices > 1) {
|
|
const loads = Array<number>(ordinarySlices).fill(0);
|
|
const longLanes = ordinarySlices - Number(registered.length < ordinary.length);
|
|
const registeredFiles = new Set(registered.map(files => files[0]));
|
|
const byWall = (a: string[], b: string[]) =>
|
|
resolvePaidShardTimeoutMs(b, opts.timeoutMs) - resolvePaidShardTimeoutMs(a, opts.timeoutMs) ||
|
|
(a[0] < b[0] ? -1 : a[0] > b[0] ? 1 : 0);
|
|
for (const files of [...registered].sort(byWall).concat(
|
|
ordinary.filter(files => !registeredFiles.has(files[0])))) {
|
|
const lanes = registeredFiles.has(files[0]) ? longLanes : ordinarySlices;
|
|
const laneKeys = (index: number) => [...allocations].filter(([, lane]) => lane === index + 1).map(([key]) => key);
|
|
// A trial whose siblings already hold every long lane may use any
|
|
// ordinary lane: independent runners outrank long-lane ownership.
|
|
const pick = (limit: number) => {
|
|
let best = -1;
|
|
for (let index = 0; index < limit; index++) {
|
|
if (sharesPanel(laneKeys(index), files[0]!)) continue;
|
|
if (best < 0 || loads[index] < loads[best]) best = index;
|
|
}
|
|
return best;
|
|
};
|
|
let lane = pick(lanes);
|
|
if (lane < 0) lane = pick(ordinarySlices);
|
|
if (lane < 0) lane = loads.slice(0, lanes).indexOf(Math.min(...loads.slice(0, lanes)));
|
|
allocations.set(files[0], lane + 1);
|
|
loads[lane] += resolvePaidShardTimeoutMs(files, opts.timeoutMs);
|
|
}
|
|
}
|
|
let ordinaryIndex = 0;
|
|
for (const files of ordinary) {
|
|
if (allocations.has(files[0])) continue;
|
|
// Round-robin, skipping a lane that already holds a trial of the same panel.
|
|
let lane = ordinaryIndex % ordinarySlices;
|
|
for (let step = 0; step < ordinarySlices; step++) {
|
|
const candidate = (ordinaryIndex + step) % ordinarySlices;
|
|
if (!sharesPanel([...allocations].filter(([, l]) => l === candidate + 1).map(([key]) => key), files[0]!)) { lane = candidate; break; }
|
|
}
|
|
ordinaryIndex++;
|
|
allocations.set(files[0], lane + 1);
|
|
}
|
|
const packed = packByRecordedDuration();
|
|
function packByRecordedDuration(): Map<string, number> | null {
|
|
const recorded = opts.durations ?? loadPaidTestDurations(rootDir, opts.tier);
|
|
if (ordinarySlices < 2 || ordinary.length === 0 || Object.keys(recorded).length === 0) return null;
|
|
const bound = (files: string[], jobs: number) => paidShardWallUpperBoundMs([...files].sort(), jobs, opts.timeoutMs);
|
|
const lanes = Array.from({ length: ordinarySlices }, (_, lane) =>
|
|
ordinary.filter(files => allocations.get(files[0]) === lane + 1).map(files => files[0]));
|
|
const caps = SUPERVISED_WORKER_COUNTS.map(jobs => Math.max(...lanes.map(files => bound(files, jobs))));
|
|
const fits = (files: string[]) => SUPERVISED_WORKER_COUNTS.every((jobs, k) => bound(files, jobs) <= caps[k]);
|
|
const known = ordinary.map(files => recordedShardMs(recorded, files[0]!))
|
|
.filter((ms): ms is number => ms !== undefined).sort((a, b) => a - b);
|
|
const fallback = known.length ? known[Math.min(known.length - 1, Math.floor(known.length * 0.75))] : 1;
|
|
const weight = (file: string) => recordedShardMs(recorded, file) ?? fallback;
|
|
const load = (files: string[]) => files.reduce((sum, file) => sum + weight(file), 0);
|
|
const registeredFiles = new Set(registered.map(files => files[0]));
|
|
// Local search from the supervised baseline: move or swap a file out of
|
|
// the heaviest slice whenever that lowers its recorded load without
|
|
// making the other slice the new maximum or breaching any worst-case cap.
|
|
// Registered files only trade places with registered files, so the long
|
|
// lanes keep their ownership.
|
|
for (let step = 0; step < 10 * ordinary.length; step++) {
|
|
const loads = lanes.map(load);
|
|
const heavy = loads.indexOf(Math.max(...loads));
|
|
let best: { gain: number; apply: () => void } | null = null;
|
|
for (let other = 0; other < lanes.length; other++) {
|
|
if (other === heavy) continue;
|
|
for (const a of lanes[heavy]) {
|
|
const moves: Array<string | null> = registeredFiles.has(a) ? lanes[other].filter(b => registeredFiles.has(b)) : [null, ...lanes[other].filter(b => !registeredFiles.has(b))];
|
|
for (const b of moves) {
|
|
const delta = weight(a) - (b === null ? 0 : weight(b));
|
|
if (delta <= 0 || loads[other] + delta >= loads[heavy]) continue;
|
|
const gain = Math.min(delta, loads[heavy] - loads[other] - delta);
|
|
if (best && gain <= best.gain) continue;
|
|
const heavyAfter = lanes[heavy].filter(file => file !== a).concat(b === null ? [] : [b]);
|
|
const otherAfter = lanes[other].filter(file => file !== b).concat([a]);
|
|
if (!fits(heavyAfter) || !fits(otherAfter)) continue;
|
|
if (sharesPanel(otherAfter, a) || (b !== null && sharesPanel(heavyAfter, b))) continue;
|
|
const [h, o] = [heavy, other];
|
|
best = { gain, apply: () => { lanes[h] = heavyAfter; lanes[o] = otherAfter; } };
|
|
}
|
|
}
|
|
}
|
|
if (!best) break;
|
|
best.apply();
|
|
}
|
|
return new Map(lanes.flatMap((files, lane) => files.map(file => [file, lane + 1] as const)));
|
|
}
|
|
runnable.forEach((files) => {
|
|
const slice = files.some(isOverlayTestFile) ? overlaySlice : (packed ?? allocations).get(files[0])!;
|
|
entries.push({ file: files[0], slice, status: 'planned', ...extras(files[0]!),
|
|
...(FILE_RETRY_BUDGETS.some(budget => budget.file === shardFile(files[0]!))
|
|
? { budget: resolvePaidShardBudget(files, opts.timeoutMs) } : {}) });
|
|
});
|
|
for (const s of skipped) entries.push({ file: s.files[0], slice: 0, status: 'skipped-by-diff', reason: s.reason, ...extras(s.files[0]!) });
|
|
for (const e of excluded) entries.push({ file: e.file, slice: 0, status: 'excluded', reason: e.reason });
|
|
entries.sort((a, b) => (a.file < b.file ? -1 : 1));
|
|
|
|
const manifest: PaidRunManifest = {
|
|
version: 1,
|
|
tier: opts.tier,
|
|
evalsAll: opts.evalsAll,
|
|
sliceCount,
|
|
selectionReason: cases.reason,
|
|
profile,
|
|
selection: cases.selection,
|
|
...(cases.coverage ? { prCoverage: cases.coverage } : {}),
|
|
entries,
|
|
};
|
|
return parseRunManifest(JSON.stringify(manifest));
|
|
}
|
|
|
|
export function parseRunManifest(raw: string): PaidRunManifest {
|
|
const parsed = JSON.parse(raw) as PaidRunManifest;
|
|
if (parsed.version !== 1) throw new Error(`unsupported manifest version: ${(parsed as { version?: unknown }).version}`);
|
|
if (!PAID_TIERS.includes(parsed.tier)) throw new Error(`manifest tier invalid: ${parsed.tier}`);
|
|
if (parsed.profile !== undefined && parsed.profile !== 'pr' && parsed.profile !== 'full') throw new Error('manifest profile invalid');
|
|
if (parsed.selection !== undefined) {
|
|
for (const [key, inventory] of [['e2e', E2E_TOUCHFILES], ['judges', LLM_JUDGE_TOUCHFILES]] as const) {
|
|
const ids = parsed.selection?.[key];
|
|
if (ids !== null && (!Array.isArray(ids) || ids.some(id => typeof id !== 'string' || !Object.hasOwn(inventory, id)) || new Set(ids).size !== ids.length)) {
|
|
throw new Error(`manifest ${key} selection invalid`);
|
|
}
|
|
}
|
|
}
|
|
if (parsed.profile === 'pr') {
|
|
const coverage = parsed.prCoverage;
|
|
if (parsed.tier !== 'gate' || !parsed.selection || !coverage ||
|
|
!['pr', 'dependents', 'full-fallback'].includes(coverage.mode) || !Array.isArray(coverage.deferred) ||
|
|
!Array.isArray(coverage.unknownFiles) || !Array.isArray(coverage.missingCoverage) ||
|
|
!Array.isArray(coverage.deferredPromptFiles) || coverage.deferredPromptFiles.some(file => typeof file !== 'string') ||
|
|
!Array.isArray(coverage.e2e) || !Array.isArray(coverage.judges) ||
|
|
coverage.unknownFiles.some(file => typeof file !== 'string') ||
|
|
coverage.needsFullValidation !== false || coverage.missingCoverage.length !== 0 ||
|
|
JSON.stringify(parsed.selection.e2e) !== JSON.stringify(coverage.e2e) ||
|
|
JSON.stringify(parsed.selection.judges) !== JSON.stringify(coverage.judges)) {
|
|
throw new Error('manifest PR coverage/selection invalid or requires full validation');
|
|
}
|
|
if (coverage.mode === 'pr' && coverage.e2e.some(id => !prProfileCaseAllowed(id, coverage.directCases ?? []))) {
|
|
throw new Error('manifest PR selection contains a broad-only case');
|
|
}
|
|
if (coverage.deferred.some(item => !Object.hasOwn(E2E_TOUCHFILES, item.id) || E2E_TIERS[item.id] !== item.tier || typeof item.reason !== 'string')) {
|
|
throw new Error('manifest deferred case is outside the broad census');
|
|
}
|
|
}
|
|
if (!Number.isInteger(parsed.sliceCount) || parsed.sliceCount <= 0) throw new Error('manifest sliceCount invalid');
|
|
if (!Array.isArray(parsed.entries)) throw new Error('manifest entries missing');
|
|
for (const entry of parsed.entries) {
|
|
if (typeof entry.file !== 'string' || !Number.isInteger(entry.slice)) throw new Error('manifest entry malformed');
|
|
if (!['planned', 'skipped-by-diff', 'excluded'].includes(entry.status)) throw new Error(`manifest entry status invalid: ${entry.status}`);
|
|
if (entry.status === 'planned' && (entry.slice < 1 || entry.slice > parsed.sliceCount)) {
|
|
throw new Error(`planned entry ${entry.file} has out-of-range slice ${entry.slice}`);
|
|
}
|
|
if (entry.estimatedMs !== undefined && (entry.status !== 'planned' || !Number.isSafeInteger(entry.estimatedMs) || entry.estimatedMs < 0)) {
|
|
throw new Error(`manifest entry ${entry.file} has an invalid estimate`);
|
|
}
|
|
if (entry.status === 'planned' && parsed.prCoverage?.mode === 'pr' && !prProfileFileSelected(entry.file, parsed.selection!, entry.excludeCases ?? [])) {
|
|
throw new Error(`manifest file is outside its PR case selection: ${entry.file}`);
|
|
}
|
|
}
|
|
for (const entry of parsed.entries) {
|
|
const caseId = shardCaseId(entry.file);
|
|
const trial = shardTrial(entry.file);
|
|
if (trial !== null) {
|
|
const plan = entry.trial;
|
|
const expected = plan && ['rule', 'behavior', 'judge'].includes(plan.kind) && typeof plan.quarantined === 'boolean'
|
|
? caseTrialPlan(caseId!, { [caseId!]: plan.kind }, plan.quarantined ? { [caseId!]: true } : {}) : undefined;
|
|
if (!Object.hasOwn(E2E_TOUCHFILES, caseId!) || !E2E_TOUCHFILES[caseId!]!.includes(shardFile(entry.file))
|
|
|| !plan || !expected || !isIsolatedCase(expected) || !sameTrialPlan(plan, expected) || trial > plan.panel.n) {
|
|
throw new Error(`Trial shard must name a registered isolated case of its file with its fixed policy panel: ${entry.file}`);
|
|
}
|
|
} else if (entry.trial !== undefined) {
|
|
throw new Error(`Only trial shards carry a trial plan: ${entry.file}`);
|
|
} else if (caseId === null ? entry.status === 'planned' && CASE_SHARDED_FILES.includes(shardFile(entry.file))
|
|
: !CASE_SHARDED_FILES.includes(shardFile(entry.file)) || !(caseId in E2E_TOUCHFILES)) {
|
|
throw new Error(`Case-sharded files plan one registered case per shard: ${entry.file}`);
|
|
}
|
|
if (entry.excludeCases !== undefined && (caseId !== null || !Array.isArray(entry.excludeCases) || entry.excludeCases.length === 0
|
|
|| entry.excludeCases.some(id => !parsed.entries.some(other => shardTrial(other.file) !== null
|
|
&& shardFile(other.file) === shardFile(entry.file) && shardCaseId(other.file) === id)))) {
|
|
throw new Error(`A file shard may exclude only cases that run as its trial shards: ${entry.file}`);
|
|
}
|
|
if (caseId !== null && entry.status === 'planned' && parsed.selection?.e2e && !parsed.selection.e2e.includes(caseId)) {
|
|
throw new Error(`Planned case shard is outside the manifest selection: ${entry.file}`);
|
|
}
|
|
}
|
|
// Panels are whole: exactly n trial entries per isolated case with one plan
|
|
// and one status, and planned trials on distinct slices when the ordinary
|
|
// slices allow it.
|
|
const panels = new Map<string, ManifestEntry[]>();
|
|
for (const entry of parsed.entries) {
|
|
const panel = trialPanelKey(entry.file);
|
|
if (panel !== null) panels.set(panel, [...(panels.get(panel) ?? []), entry]);
|
|
}
|
|
const reservedOverlay = parsed.sliceCount > 1 && parsed.entries.some(entry => entry.status === 'planned' && isOverlayTestFile(entry.file));
|
|
const ordinarySliceCount = parsed.sliceCount - Number(reservedOverlay);
|
|
for (const [panel, trials] of panels) {
|
|
const n = trials[0]!.trial!.panel.n;
|
|
const indices = trials.map(entry => shardTrial(entry.file)!).sort((a, b) => a - b);
|
|
if (indices.length !== n || indices.some((index, i) => index !== i + 1)
|
|
|| trials.some(entry => !sameTrialPlan(entry.trial, trials[0]!.trial) || entry.status !== trials[0]!.status)
|
|
|| parsed.entries.some(entry => normalizeRelativePath(entry.file) === panel)) {
|
|
throw new Error(`Isolated case ${panel} must plan exactly its ${n} trials together`);
|
|
}
|
|
const slices = trials.filter(entry => entry.status === 'planned').map(entry => entry.slice);
|
|
if (ordinarySliceCount >= n && new Set(slices).size !== slices.length) {
|
|
throw new Error(`Trials of ${panel} share a slice; each trial needs its own runner`);
|
|
}
|
|
}
|
|
if (parsed.prCoverage?.mode === 'pr') {
|
|
const planned = parsed.entries.filter(entry => entry.status === 'planned').map(entry => normalizeRelativePath(entry.file));
|
|
const required: string[][] = Object.entries(prProfileFileMap(parsed.selection!.e2e)).flatMap(([file, ids]) => {
|
|
const selected = ids.filter(id => parsed.selection!.e2e!.includes(id));
|
|
if (!selected.length) return [];
|
|
const owners = new Set(selected.flatMap(id => {
|
|
const trials = parsed.entries.filter(entry => shardTrial(entry.file) !== null && shardFile(entry.file) === file && shardCaseId(entry.file) === id);
|
|
if (trials.length) return trials.map(entry => normalizeRelativePath(entry.file));
|
|
return [CASE_SHARDED_FILES.includes(file) ? `${file}#${id}` : file];
|
|
}));
|
|
return [[...owners]];
|
|
});
|
|
if (parsed.selection!.judges!.length) required.push(['test/skill-llm-eval.test.ts']);
|
|
for (const owners of required) {
|
|
if (owners.some(owner => planned.filter(key => key === owner).length !== 1)
|
|
|| planned.filter(key => shardFile(key) === shardFile(owners[0]!)).length !== owners.length) {
|
|
throw new Error(`PR selected cases require exactly one planned owning file: ${owners.join(', ')}`);
|
|
}
|
|
}
|
|
}
|
|
if (parsed.plan !== undefined) {
|
|
const plan = parsed.plan;
|
|
const count = (value: unknown) => Number.isSafeInteger(value) && Number(value) > 0;
|
|
if (!plan || typeof plan !== 'object' || !count(plan.sliceBudgetMs) || !count(plan.jobs) || !count(plan.ciTimeoutMinutes)
|
|
|| !Array.isArray(plan.estimatedSliceMs) || plan.estimatedSliceMs.length !== parsed.sliceCount
|
|
|| !plan.estimatedSliceMs.every(ms => Number.isSafeInteger(ms) && ms >= 0)
|
|
|| parsed.entries.some(entry => entry.status === 'planned' && entry.estimatedMs === undefined)
|
|
|| (plan.sliceCiTimeoutMinutes !== undefined && (!Array.isArray(plan.sliceCiTimeoutMinutes)
|
|
|| plan.sliceCiTimeoutMinutes.length !== parsed.sliceCount || !plan.sliceCiTimeoutMinutes.every(count)
|
|
|| Math.max(...plan.sliceCiTimeoutMinutes) !== plan.ciTimeoutMinutes))) {
|
|
throw new Error('manifest slice plan malformed');
|
|
}
|
|
const codexSlices = codexSlicesOf(parsed.tier, parsed.entries);
|
|
if (JSON.stringify(plan.codexSlices ?? []) !== JSON.stringify(codexSlices)) {
|
|
throw new Error(`manifest Codex slices must be exactly the slices holding Codex shards: ${JSON.stringify(codexSlices)}`);
|
|
}
|
|
const mixed = parsed.entries.find(entry => entry.status === 'planned' && codexSlices.includes(entry.slice) && !isCodexShard(entry.file, parsed.tier));
|
|
if (mixed) throw new Error(`Codex slice ${mixed.slice} also holds ${mixed.file}; only Codex shards run on the host-run Codex job`);
|
|
}
|
|
const keys = parsed.entries.map(entry => normalizeRelativePath(entry.file));
|
|
if (new Set(keys).size !== keys.length) throw new Error('Duplicate manifest entry');
|
|
// Unique result slugs: shard artifacts merge by path, so a shared slug would
|
|
// let one trial's records overwrite another's.
|
|
const slugs = new Map<string, string>();
|
|
for (const entry of parsed.entries) {
|
|
const slug = shardSlug([entry.file]);
|
|
if (slugs.has(slug)) throw new Error(`Shards ${slugs.get(slug)} and ${entry.file} share the result slug ${slug}`);
|
|
slugs.set(slug, entry.file);
|
|
}
|
|
const overlaySlice = parsed.sliceCount;
|
|
const plannedOverlays = parsed.entries.filter(entry => entry.status === 'planned' && isOverlayTestFile(entry.file));
|
|
if (plannedOverlays.some(entry => entry.slice !== overlaySlice)) {
|
|
throw new Error('Overlay manifest entries must share the final ordinary slice to preserve one-process API admission');
|
|
}
|
|
if (plannedOverlays.length && overlaySlice > 1 && parsed.entries.some(entry =>
|
|
entry.status === 'planned' && !isOverlayTestFile(entry.file) && entry.slice === overlaySlice)) {
|
|
throw new Error('The final ordinary manifest slice is reserved for overlay files');
|
|
}
|
|
for (const budget of FILE_RETRY_BUDGETS) {
|
|
const entries = parsed.entries.filter(entry => shardFile(entry.file) === budget.file);
|
|
const fileKeys = entries.filter(entry => shardCaseId(entry.file) === null).length;
|
|
const caseKeys = entries.filter(entry => shardCaseId(entry.file) !== null && shardTrial(entry.file) === null).length;
|
|
if (fileKeys > 1 || (fileKeys === 1 && caseKeys > 0)) throw new Error(`Duplicate registered manifest entry: ${budget.file}`);
|
|
for (const entry of entries.filter(entry => entry.status === 'planned')) {
|
|
if (!entry.budget) throw new Error(`Registered manifest needs an explicit budget record: ${budget.file}`);
|
|
const expected = resolvePaidShardBudget([entry.file], entry.budget.source === 'explicit' ? entry.budget.timeoutMs : undefined);
|
|
if (!sameBudget(entry.budget, expected)) throw new Error(`Registered manifest budget differs from declared policy: ${budget.file}`);
|
|
}
|
|
}
|
|
return parsed;
|
|
}
|
|
|
|
export interface SliceResult {
|
|
version: 1;
|
|
tier: PaidTier;
|
|
profile?: PaidProfile;
|
|
selection?: PaidCaseSelection;
|
|
sliceIndex: number;
|
|
sliceCount: number;
|
|
timeoutOverrideMs?: number;
|
|
/** CI run attempt (github.run_attempt) that produced this slice; absent means 1. */
|
|
attempt?: number;
|
|
/** Epoch ms bounds of the slice's shard execution (lane wall time). */
|
|
startedAt?: number;
|
|
finishedAt?: number;
|
|
/** A progress checkpoint the executor never replaced: the job ended (ceiling or cancellation) before its final result. */
|
|
checkpoint?: true;
|
|
outcomes: Array<Pick<ShardOutcome, 'files' | 'status' | 'exitCode' | 'elapsedMs' | 'executedTests' | 'skippedTests' | 'budget' | 'reused' | 'runnerError' | 'trial' | 'inputKey' | 'sliceDeadline'>>;
|
|
}
|
|
|
|
/**
|
|
* Slice exit = execution completeness, never the semantic verdict. A rule
|
|
* shard that did not pass fails the slice (unchanged fail-closed rule); an
|
|
* isolated trial shard fails it only when the harness produced no trial
|
|
* record. Failed, timed-out or crashed trials are verdict input for the
|
|
* report's panelVerdict(), so a 2/3 PASS panel never reds its runner.
|
|
*/
|
|
export function sliceExitCode(outcomes: ReadonlyArray<Pick<ShardOutcome, 'status' | 'trial'>>): number {
|
|
return outcomes.every(outcome => outcome.trial !== undefined
|
|
? outcome.trial.outcome !== null
|
|
: outcome.status === 'passed' || outcome.status === 'skipped-by-diff') ? 0 : 1;
|
|
}
|
|
|
|
/** A hollow-guarded trial shard has no trial record: the guard's verdict is a harness problem. */
|
|
export function guardTrialRecords<T extends Pick<ShardOutcome, 'status' | 'trial'>>(outcomes: T[]): T[] {
|
|
return outcomes.map(outcome => outcome.trial && outcome.status === 'passed-empty' && outcome.trial.outcome !== null
|
|
? { ...outcome, trial: { ...outcome.trial, outcome: null, harness: 'hollow: executed no case' } } : outcome);
|
|
}
|
|
|
|
/**
|
|
* Reconcile slice results against the manifest — the fail-closed aggregation.
|
|
* Problems (any → non-zero): a slice index missing entirely (a cancelled or
|
|
* crashed executor whose artifact never landed), a planned entry no slice
|
|
* reported, an entry reported by the wrong/duplicate slice, or any reported
|
|
* outcome that is not a pass.
|
|
*/
|
|
export function verifySliceResults(
|
|
manifest: PaidRunManifest,
|
|
results: SliceResult[],
|
|
): { ok: boolean; problems: string[] } {
|
|
const problems: string[] = [];
|
|
try { parseRunManifest(JSON.stringify(manifest)); }
|
|
catch (error) { problems.push(`Invalid run manifest: ${error instanceof Error ? error.message : String(error)}`); }
|
|
const byIndex = new Map<number, SliceResult>();
|
|
for (const result of results) {
|
|
if (result.version !== 1) { problems.push(`slice result with unsupported version: ${String(result.version)}`); continue; }
|
|
if (result.tier !== manifest.tier) problems.push(`slice ${result.sliceIndex} ran tier ${result.tier}, manifest says ${manifest.tier}`);
|
|
if (manifest.profile === 'pr' && (result.profile !== 'pr' || JSON.stringify(result.selection) !== JSON.stringify(manifest.selection))) {
|
|
problems.push(`slice ${result.sliceIndex} did not bind the manifest PR case selection`);
|
|
}
|
|
if (byIndex.has(result.sliceIndex)) problems.push(`duplicate result for slice ${result.sliceIndex}`);
|
|
if (result.checkpoint) problems.push(`slice ${result.sliceIndex}/${result.sliceCount} ended before its final result (job ceiling or cancellation); its outcomes come from the last checkpoint: TIMEOUT`);
|
|
byIndex.set(result.sliceIndex, result);
|
|
}
|
|
for (let index = 1; index <= manifest.sliceCount; index += 1) {
|
|
if (!byIndex.has(index)) problems.push(`slice ${index}/${manifest.sliceCount} reported NO result — cancelled/crashed executor, not a pass`);
|
|
}
|
|
|
|
const reported = new Map<string, { slice: number; status: ShardStatus; trial?: ShardTrialRecord; sliceDeadline?: ShardOutcome['sliceDeadline'] }>();
|
|
const planned = new Map(manifest.entries.map(entry => [normalizeRelativePath(entry.file), entry]));
|
|
for (const result of results) {
|
|
for (const outcome of result.outcomes) {
|
|
if (outcome.files.some(file => FILE_RETRY_BUDGETS.some(budget => budget.file === shardFile(file)) || shardCaseId(file) !== null) && outcome.files.length !== 1) {
|
|
problems.push('Registered result must report its own shard');
|
|
}
|
|
const file = normalizeRelativePath(outcome.files[0] ?? '');
|
|
const trialPlan = planned.get(file)?.trial;
|
|
if (manifest.prCoverage?.mode === 'pr') {
|
|
const expected = expectedPrCaseCount(file, manifest.selection!, planned.get(file)?.excludeCases);
|
|
const executed = outcome.executedTests === null || outcome.skippedTests === null
|
|
? -1 : outcome.executedTests - outcome.skippedTests;
|
|
// A trial's exit status is its verdict (panelVerdict decides); its
|
|
// completeness is the trial record checked below.
|
|
if ((!trialPlan && outcome.exitCode !== 0) || expected < 1 || (!trialPlan && executed !== expected)) {
|
|
problems.push(`PR profile expected ${expected} executed cases in ${file}, received ${executed}`);
|
|
}
|
|
}
|
|
if (trialPlan) {
|
|
const t = outcome.trial;
|
|
if (!t || t.case !== shardCaseId(file) || t.trial !== shardTrial(file) || !sameTrialPlan(t, trialPlan)
|
|
|| !(t.outcome === null || ['passed', 'failed', 'skipped'].includes(t.outcome))
|
|
|| (t.outcome === 'failed') !== (t.failure_class !== undefined)) {
|
|
problems.push(`${file}: trial record missing or does not match its planned trial`);
|
|
}
|
|
} else if (outcome.trial !== undefined) {
|
|
problems.push(`${file}: an unplanned trial record`);
|
|
}
|
|
if (shardCaseId(file) !== null && outcome.status === 'passed'
|
|
&& (outcome.executedTests === null || outcome.skippedTests === null || outcome.executedTests - outcome.skippedTests !== 1)) {
|
|
problems.push(`Case shard must execute exactly its one case: ${file}`);
|
|
}
|
|
if (outcome.reused !== undefined) {
|
|
const r = outcome.reused;
|
|
if (manifest.prCoverage?.mode !== 'pr') problems.push(`${file}: only the fast PR profile may reuse results; this lane executes fresh`);
|
|
const reusedVerdictOk = outcome.trial ? outcome.trial.outcome !== null : outcome.status === 'passed' && outcome.exitCode === 0;
|
|
if (!reusedVerdictOk || !/^[a-f0-9]{64}$/.test(r?.inputKey ?? '')
|
|
|| !/^[\w./-]{1,160}$/.test(r?.runId ?? '') || !/^[a-f0-9]{40}$/.test(r?.revision ?? '') || !Number.isSafeInteger(r?.completedAt) || r.completedAt <= 0) {
|
|
problems.push(`${file}: malformed reused result`);
|
|
}
|
|
}
|
|
if (outcome.inputKey !== undefined && !/^[a-f0-9]{64}$/.test(outcome.inputKey)) problems.push(`${file}: malformed input identity`);
|
|
if (reported.has(file)) problems.push(`${file} reported by two slices`);
|
|
reported.set(file, { slice: result.sliceIndex, status: outcome.status, ...(outcome.trial ? { trial: outcome.trial } : {}),
|
|
...(outcome.sliceDeadline ? { sliceDeadline: outcome.sliceDeadline } : {}) });
|
|
const registered = FILE_RETRY_BUDGETS.find(budget => budget.file === shardFile(file));
|
|
const finding = STRICT_RETRY_CASE_BUDGETS.find(budget => budget.file === file);
|
|
if (finding) {
|
|
// Full-census runs must account for every registered case. A manifest
|
|
// explicitly marked selective may report its executed subset.
|
|
if (outcome.exitCode !== 0 || !Number.isInteger(outcome.executedTests) ||
|
|
outcome.executedTests! < 1 || outcome.executedTests! > finding.cases ||
|
|
(manifest.evalsAll !== false && outcome.executedTests !== finding.cases) || outcome.skippedTests !== 0) {
|
|
problems.push(`Finding workflow must execute real unskipped cases with exit zero: ${file}`);
|
|
}
|
|
}
|
|
if (registered) {
|
|
try {
|
|
const planned = manifest.entries.find(entry => normalizeRelativePath(entry.file) === file)?.budget;
|
|
const expected = resolvePaidShardBudget([file], result.timeoutOverrideMs ??
|
|
(planned?.source === 'explicit' ? planned.timeoutMs : undefined));
|
|
if (!sameBudget(outcome.budget, expected)) problems.push(`Registered effective result budget differs from its planned/explicit allocation: ${file}`);
|
|
} catch { problems.push(`Invalid registered effective result budget: ${file}`); }
|
|
}
|
|
}
|
|
}
|
|
for (const entry of manifest.entries) {
|
|
if (entry.status !== 'planned') continue;
|
|
const got = reported.get(normalizeRelativePath(entry.file));
|
|
if (!got) {
|
|
if (byIndex.has(entry.slice)) problems.push(`planned ${entry.file} (slice ${entry.slice}) was never reported`);
|
|
continue; // the missing-slice problem above already covers it
|
|
}
|
|
if (got.slice !== entry.slice) problems.push(`${entry.file} planned for slice ${entry.slice} but reported by slice ${got.slice}`);
|
|
// Isolated trial shards: harness health only; the panel verdict gates.
|
|
if (entry.trial) {
|
|
if (got.trial?.outcome === null) problems.push(`${entry.file}: no trial record (${got.trial.harness ?? 'unknown'})`);
|
|
} else if (got.status !== 'passed') {
|
|
// The slice deadline distinguishes work that never started (INFRA) from work killed in flight (TIMEOUT).
|
|
const why = got.sliceDeadline === 'not_run' ? ' (not_run: the slice deadline passed before it started)'
|
|
: got.sliceDeadline === 'hung' ? ' (hung: killed at the slice deadline)' : '';
|
|
problems.push(`${entry.file}${why}: ${got.status}`);
|
|
}
|
|
}
|
|
return { ok: problems.length === 0, problems };
|
|
}
|
|
|
|
/** Budget-mode plan lines: every slice with its estimate, files and retries. */
|
|
export function formatSlicePlan(manifest: PaidRunManifest): string[] {
|
|
const plan = manifest.plan;
|
|
if (!plan) return [];
|
|
const minutes = (ms: number) => (ms / 60_000).toFixed(1);
|
|
const lines = [`[test:paid] slice plan: ${manifest.sliceCount} slice(s) x ${plan.jobs} worker(s), budget ${minutes(plan.sliceBudgetMs)}m per slice, `
|
|
+ `longest estimate ${minutes(Math.max(0, ...plan.estimatedSliceMs))}m, CI job timeout ${plan.ciTimeoutMinutes}m (largest per-slice ceiling)`];
|
|
for (let slice = 1; slice <= manifest.sliceCount; slice++) {
|
|
const mine = sliceExecutionOrder(manifest.entries.filter(entry => entry.status === 'planned' && entry.slice === slice));
|
|
const over = plan.estimatedSliceMs[slice - 1]! > plan.sliceBudgetMs ? ' [over budget: longer than one runner allows]' : '';
|
|
const ceiling = plan.sliceCiTimeoutMinutes?.[slice - 1];
|
|
lines.push(` slice ${slice}: ~${minutes(plan.estimatedSliceMs[slice - 1]!)}m${ceiling !== undefined ? `, job ceiling ${ceiling}m` : ''}${over}`);
|
|
for (const entry of mine) lines.push(` ${entry.file} ~${minutes(entry.estimatedMs ?? 0)}m retries=${retriesForFiles([entry.file])}`);
|
|
}
|
|
return lines;
|
|
}
|
|
|
|
/**
|
|
* Capacity preflight (E-A10): what the plan asks of the runner pool and the
|
|
* API. Sessions are planned shard processes; at most jobs of them run per
|
|
* slice at once. Waves > 1 mean slices queue behind the matrix cap and the
|
|
* lane wall grows by whole slices.
|
|
*/
|
|
export function formatCapacityPreflight(manifest: PaidRunManifest, maxParallel?: number): string[] {
|
|
const planned = manifest.entries.filter(entry => entry.status === 'planned');
|
|
const trials = planned.filter(entry => entry.trial);
|
|
const jobs = manifest.plan?.jobs ?? DEFAULT_JOBS;
|
|
const longest = [...trials].sort((a, b) => (b.estimatedMs ?? 0) - (a.estimatedMs ?? 0))[0];
|
|
const waves = maxParallel ? Math.ceil(manifest.sliceCount / maxParallel) : null;
|
|
return [
|
|
`[test:paid] capacity: ${manifest.sliceCount} slice(s), ${planned.length} planned shard(s) (${planned.length - trials.length} rule/judge, ${trials.length} trial shard(s) in ${new Set(trials.map(entry => trialPanelKey(entry.file))).size} panel(s)); `
|
|
+ `peak ${Math.min(manifest.sliceCount, maxParallel ?? manifest.sliceCount) * jobs} concurrent shard process(es)`
|
|
+ (waves !== null ? `; wave(s) at max-parallel ${maxParallel}: ${waves}` : ''),
|
|
...(longest ? [`[test:paid] capacity: longest indivisible trial ~${((longest.estimatedMs ?? 0) / 60_000).toFixed(1)}m (${longest.file})`] : []),
|
|
...(waves !== null && waves > 1 ? [`[test:paid] capacity: ⚠ ${manifest.sliceCount} slices exceed max-parallel ${maxParallel}; later slices queue for a second wave`] : []),
|
|
];
|
|
}
|
|
|
|
export function formatProfileCoverage(manifest: PaidRunManifest): string[] {
|
|
const coverage = manifest.prCoverage;
|
|
return [
|
|
`[test:paid] coverage: profile=${manifest.profile ?? 'full'} mode=${coverage?.mode ?? 'full'}; selected E2E=${manifest.selection?.e2e?.length ?? 'all'}, judges=${manifest.selection?.judges?.length ?? 'all'}`,
|
|
...(coverage ? [`[test:paid] deferred: ${coverage.deferred.length} broad behaviors, ${coverage.deferredPromptFiles.length} changed prompts without quick live coverage; these are not PR passes`] : []),
|
|
];
|
|
}
|
|
|
|
/** Every collector record counts: paid evals never retry, so a later record never replaces an earlier one. */
|
|
export function collectorOutcomeCounts(results: Array<{ tests?: Array<{
|
|
name: string; suite?: string; passed: boolean; execution?: string; manual_review?: unknown;
|
|
}> }>): { executed: number; reused: number; passed: number; failed: number; manual_accepted: number; attempts: number } {
|
|
const counts = { executed: 0, reused: 0, passed: 0, failed: 0, manual_accepted: 0, attempts: 0 };
|
|
for (const result of results) {
|
|
for (const entry of result.tests ?? []) {
|
|
if (!entry || typeof entry !== 'object' || typeof entry.name !== 'string' || typeof entry.passed !== 'boolean') continue;
|
|
counts.attempts++;
|
|
counts[entry.execution === 'reused' ? 'reused' : 'executed']++;
|
|
const outcome = evalEntryOutcome(entry);
|
|
counts[outcome === 'manual-review' ? 'manual_accepted' : outcome]++;
|
|
}
|
|
}
|
|
return counts;
|
|
}
|