PulsePlate Playwright E2E

When to use

• •Step 3 extension after core productivity pack is in place.
• •Validating key browser flows in frontend/ (login, onboarding, premium gates, exports).
• •Reproducing UI bugs that are hard to isolate with unit tests only.
• •Running predefined Step 3 smoke scenarios from the runbook.

Inputs required

• •Target environment URL (local or preview).
• •Flow list (1-3 user journeys to validate).
• •Expected pass criteria per journey.
• •Scenario IDs (from runbook matrix, e.g. E2E-01 ... E2E-04).

Procedure (commands)

1. •

   Ensure frontend dependencies are ready:

   bash

   cd frontend
   npm ci
   cd ..
   

2. •

   Use Playwright skill/tooling to run browser automation against selected flows.

3. •

   Capture deterministic artifacts:

   • •command and config used
   • •failing step and selector/action
   • •screenshot path or trace path when available
4. •

   Re-run only failing flow after fix.

5. •

   Follow the runbook evidence contract for every flow.

Output format

• •Flow matrix: flow name + pass/fail.
• •Scenario IDs: include exact executed scenario IDs.
• •Failure evidence: raw failing lines and failing step.
• •Pointers: file references for impacted UI/API contracts.
• •Fix plan: minimal changes to restore flow.
• •Rerun: exact command sequence.

Guardrails

• •Scope is browser E2E only; no desktop RPA.
• •Do not use Playwright to bypass thin-client policy or API contracts.
• •Keep runs targeted; avoid broad unstable suites without need.
• •Do not claim release readiness solely from E2E; keep hard backend gates mandatory.

SoT links

• •frontend/AGENTS.md
• •tools/codex_skills/pulseplate-frontend-ui/SKILL.md
• •docs/dev/PLAYWRIGHT_E2E_RUNBOOK.md
• •.cursor/agents/dev-operator.md
• •AGENTS.md