test(e2e): gate-19 186 → 161 — 25 verified anchors, 1 refusal, 2 false anchors removed

[rubenvdlinde]

gate-19: 186 → 161

Measured full scope (hydra-gates --full) with hydra-gates @ main (b8c7ead, 1735-line checker — the ref CI pins), against openbuild@development:

before  {scenarios: 767, covered: 72, excluded: 509, uncovered: 186}
after   {scenarios: 767, covered: 97, excluded: 509, uncovered: 161}

The 186 baseline reproduces run 31459739568's figure exactly.

⚠️ This PR's own gate-19 cell proves nothing about that. Gate-19 is diff-scoped; this PR changes no spec file, so its scope is empty and it will report PASS trivially. The evidence is the negative control below, which is scope-independent.

Negative control

state	uncovered
current	161
tests/e2e/builder-undo-redo.spec.ts removed	172 (+11 — exactly its 11 anchors)
restored	161

Positive control

Corrupting one slug to #adding-a-step-groups-fields-by-BOGUS dropped covered 99 → 98; restoring returned it to 99. The gate validates slugs, so the other 24 are real matches — not tags it accepted without looking.

What the 25 anchors are

Not new tests. builder-undo-redo.spec.ts and form-editor-logic.spec.ts already drive these scenarios and already run green; they carried no @e2e reference, so the gate could not see them. Every mapping was re-derived by reading the test body against the scenario text — a title that resembles a scenario is not coverage of it.

• builder-undo-redo — 11 anchors across 6 tests
• form-editor-logic — 14 anchors across 5 tests

One of the 14 needed the test extended before it could be anchored honestly: deleting-a-step-returns-its-fields-to-the-unassigned-pool asserted the removal and the pool re-appearance but not the spec's final AND (save auto-assigns an unassigned field to the last step, with a warning mark). The surface exists — assignUnassignedFieldsToFinalStep() in src/services/manifestValidation/formLogic.js, called from PageDesignerHost.vue's save() — so the test now re-adds a step, asserts the pool note, saves, and polls the persisted manifest for the assignment.

One scenario refused, with no exclusion written

form-editor-logic::raw-json-authored-logic-survives-unrelated-editor-edits asserts a Design ↔ Raw JSON tab round-trip. That tab does not exist on /builder/:slug/pages; the raw editor is a sidebar tab on the VirtualAppDetail page, writing the Application object while the designer saves onto the ApplicationVersion. Different surface, different storage target.

Anchoring it would have been false. Excluding it would have claimed "a browser cannot observe this", which is not the reason — the reason is that the spec describes a surface the product does not have. So it stays uncovered, with an in-file comment explaining why, so nobody later "fixes" it with a false anchor.

Two anchors REMOVED — this costs the number and is the right call

save-as-template.spec.ts credited viewer-cannot-save-a-template and seeded-cards-remain-read-only to a test whose own comment, three lines below the tags, says it proves neither: the assertion is toHaveCount(0) on a button, under an admin session, on a page where the button is not scoped to appear. Filed as #178.

Removing them did not raise uncovered — 161 either way. Both scenarios are also swallowed by a requirement-level @e2e exclude whose stated reason is about something else (REQ-SAT-001's reason is about "the captured record's internals"; the scenario it exempts is about whether a viewer sees a button). The false anchor was a second green layer over a scenario that was already invisibly exempt.

Chasing that produced a fleet-wide finding, filed as ConductionNL/.github#356: 471 of openbuild's 509 exclusions (92.5%) share a reason with a sibling — 42 reasons, one covering 32 scenarios, several beginning "mixed spec — …". Also worth knowing when reading any gate-19 percentage: coverage_pct is covered / (scenarios − excluded), so openbuild's headline 27.9% is 9.4% (72/767) unqualified.

Not done here

No test was skipped, no timeout widened, no mode: 'serial', no waitForLoadState('networkidle'), and no exclusion added.

ESLint on all three changed files reports only the pre-existing n/no-unpublished-import on the @playwright/test import, which every test file in the repo carries.

The red E2E Tests (Playwright) job is a separate, pre-existing failure — diagnosed and fixed in #177, which is open for review.

[github-actions]

Quality Report — ConductionNL/openbuild @ dd18303

Check	PHP	Vue	Security	License	Tests
lint	✅
phpcs	✅
phpmd	✅
psalm	✅
phpstan	✅
phpmetrics	✅
eslint		✅
stylelint		✅
build		✅
check-manifest		✅
test-l10n		✅
check-gitignore		✅
check-nc-floor		✅
composer			✅	✅ 101/101
npm			✅	✅ 654/654
PHPUnit					✅
Newman					⏭️
Playwright					❌
Hydra gates					❌

---

Quality workflow — 2026-08-11 13:48 UTC

Download the full PDF report from the workflow artifacts.

[github-actions]

Quality Report — ConductionNL/openbuild @ 40d4357

Check	PHP	Vue	Security	License	Tests
lint	✅
phpcs	✅
phpmd	✅
psalm	✅
phpstan	✅
phpmetrics	✅
eslint		✅
stylelint		✅
build		❌
check-manifest		✅
test-l10n		✅
check-gitignore		✅
check-nc-floor		✅
composer			✅	✅ 101/101
npm			✅	✅ 654/654
PHPUnit					❌
Newman					⏭️
Playwright					❌
Hydra gates					❌

---

Quality workflow — 2026-08-11 13:51 UTC

Download the full PDF report from the workflow artifacts.

[github-actions]

Quality Report — ConductionNL/openbuild @ 6f6f6d2

Check	PHP	Vue	Security	License	Tests
lint	✅
phpcs	✅
phpmd	✅
psalm	✅
phpstan	✅
phpmetrics	✅
eslint		✅
stylelint		✅
build		✅
check-manifest		✅
test-l10n		✅
check-gitignore		✅
check-nc-floor		✅
composer			✅	✅ 101/101
npm			✅	✅ 654/654
PHPUnit					✅
Newman					⏭️
Playwright					❌
Hydra gates					✅

---

Quality workflow — 2026-08-11 14:26 UTC

Download the full PDF report from the workflow artifacts.