name	e2e
description	Run the Playwright E2E suite with a clean backend lifecycle — kill stale port 8080 backend, run suite, iterate failures to green. Use for "run e2e", "fix e2e tests", E2E verification.

E2E test loop

Check port 8080: ss -ltn | grep ':8080 '. If occupied, kill it (fuser -k 8080/tcp), wait 2s, confirm free. Playwright's reuseExistingServer would otherwise reuse a stale backend and produce false failures.
Run the suite: cd frontend && npm run e2e. Playwright's webServer starts a fresh backend itself (./gradlew :backend:e2eApp, up to 180s) plus the Vite dev server — do not start a backend manually.
On failures: root-cause from test output and traces in frontend/e2e/test-results/. Fix, then re-run only the failed spec first (npm run e2e -- <spec-file>), then the full suite.
Iterate until the whole suite is green. Green full suite is the completion bar — never claim done on partial results.
If a run was interrupted and left a backend on 8080, kill it so the next run starts clean.

name	e2e
description	Run the Playwright E2E suite with a clean backend lifecycle — kill stale port 8080 backend, run suite, iterate failures to green. Use for "run e2e", "fix e2e tests", E2E verification.

Check port 8080: ss -ltn | grep ':8080 '. If occupied, kill it (fuser -k 8080/tcp), wait 2s, confirm free. Playwright's reuseExistingServer would otherwise reuse a stale backend and produce false failures.
Run the suite: cd frontend && npm run e2e. Playwright's webServer starts a fresh backend itself (./gradlew :backend:e2eApp, up to 180s) plus the Vite dev server — do not start a backend manually.
On failures: root-cause from test output and traces in frontend/e2e/test-results/. Fix, then re-run only the failed spec first (npm run e2e -- <spec-file>), then the full suite.
Iterate until the whole suite is green. Green full suite is the completion bar — never claim done on partial results.
If a run was interrupted and left a backend on 8080, kill it so the next run starts clean.

e2e