ring/dev-team/skills/dev-frontend-e2e/SKILL.md
Fred Amaral fa6c4c87e8
docs(frontmatter): create canonical schema to standardize component metadata
feat(hooks): add schema validator and enhance skill generation
refactor(*): align all components with the new frontmatter schema
refactor(commands): replace arguments object with simple argument-hint
refactor(agents): remove invalid version field from agent frontmatter
test(hooks): add unit tests for frontmatter validation and generation
2026-04-06 09:52:28 -07:00

8.1 KiB

name description trigger skip_when NOT_skip_when sequence related input_schema output_schema verification
ring:dev-frontend-e2e Gate 5 of frontend development cycle - ensures all user flows from product-designer have passing E2E tests with Playwright across browsers. - After visual testing complete (Gate 4) - MANDATORY for all frontend development tasks - Validates user flows end-to-end - Not inside a frontend development cycle (ring:dev-cycle-frontend) - Backend-only project with no UI components - Task is documentation-only, configuration-only, or non-code - No user-facing flows were added or changed in this cycle - "Unit tests cover the flow" - Unit tests don't test real browser + API interaction. - "We only need Chromium" - Users use Firefox and Safari too. - "Happy path is enough" - Error handling MUST be tested.
after before
ring:dev-frontend-visual
ring:dev-frontend-performance
complementary
ring:dev-cycle-frontend
ring:dev-frontend-visual
ring:qa-analyst-frontend
required optional
name type description
unit_id string Task or subtask identifier
name type items description
implementation_files array string Files from Gate 0 implementation
name type description
user_flows_path string Path to user-flows.md from product-designer
name type description
backend_handoff object Backend endpoints and contracts from backend dev cycle
name type description
gate4_handoff object Full handoff from Gate 4 (visual testing)
format required_sections metrics
markdown
name pattern required
E2E Testing Summary ^## E2E Testing Summary true
name pattern required
Flow Coverage ^## Flow Coverage true
name pattern required
Handoff to Next Gate ^## Handoff to Next Gate true
name type values
result enum
PASS
FAIL
name type
flows_tested integer
name type
happy_path_tests integer
name type
error_path_tests integer
name type
browsers_passed integer
name type
iterations integer
automated manual
command description success_pattern
grep -rn 'test(' --include='*.spec.ts' --include='*.e2e.ts' . Playwright tests exist test(
command description success_pattern
grep -rn 'getByRole|getByTestId|getByLabel' --include='*.spec.ts' . Semantic selectors used getByRole|getByTestId|getByLabel
All user flows from product-designer have E2E tests
Error paths tested (API 500, timeout, validation)
Tests pass on Chromium, Firefox, and WebKit
Responsive viewports covered
3 consecutive passes without flaky failures

Dev Frontend E2E Testing (Gate 5)

Overview

Ensure all user flows from ring:product-designer have passing Playwright E2E tests across Chromium, Firefox, and WebKit with responsive viewport coverage.

Core principle: If the product-designer defined a user flow, it must have an E2E test. If the user can encounter an error, it must be tested.

<block_condition>

  • Untested user flow = FAIL
  • No error path tests = FAIL
  • Fails on any browser = FAIL
  • Flaky tests (fail on consecutive runs) = FAIL </block_condition>

CRITICAL: Role Clarification

This skill ORCHESTRATES. Frontend QA Analyst Agent (e2e mode) EXECUTES.

Who Responsibility
This Skill Gather requirements, dispatch agent, track iterations
QA Analyst Frontend Agent Write Playwright tests, run cross-browser, verify flows

Standards Reference

MANDATORY: Load testing-e2e.md standards via WebFetch.

<fetch_required> https://raw.githubusercontent.com/LerianStudio/ring/main/dev-team/docs/standards/frontend/testing-e2e.md </fetch_required>


Step 1: Validate Input

REQUIRED INPUT:
- unit_id: [task/subtask being tested]
- implementation_files: [files from Gate 0]

OPTIONAL INPUT:
- user_flows_path: [path to user-flows.md]
- backend_handoff: [endpoints, contracts from backend cycle]
- gate4_handoff: [full Gate 4 output]

if any REQUIRED input is missing:
  → STOP and report: "Missing required input: [field]"

if user_flows_path provided:
  → Load user flows as E2E test scenarios
  → All flows MUST be covered

if backend_handoff provided:
  → Verify E2E tests exercise backend endpoints
  → Verify request payloads match contracts

Step 2: Dispatch Frontend QA Analyst Agent (E2E Mode)

Task tool:
  subagent_type: "ring:qa-analyst-frontend"
  prompt: |
    **MODE:** E2E TESTING (Gate 5)

    **Standards:** Load testing-e2e.md

    **Input:**
    - Unit ID: {unit_id}
    - Implementation Files: {implementation_files}
    - User Flows: {user_flows_path or "N/A"}
    - Backend Handoff: {backend_handoff or "N/A"}

    **Requirements:**
    1. Create Playwright tests for all user flows
    2. Test happy path + error paths (API 500, timeout, validation)
    3. Run on Chromium, Firefox, WebKit
    4. Test responsive viewports (mobile, tablet, desktop)
    5. Use data-testid or semantic role selectors only
    6. Run 3x consecutively to verify no flaky tests

    **Output Sections Required:**
    - ## E2E Testing Summary
    - ## Flow Coverage
    - ## Handoff to Next Gate

Step 3: Evaluate Results

Parse agent output:

if "Status: PASS" in output:
  → Gate 5 PASSED
  → Return success with metrics

if "Status: FAIL" in output:
  → If flow not covered: re-dispatch agent to add missing tests
  → If browser failure: re-dispatch implementation agent to fix
  → If flaky: re-dispatch agent to stabilize tests
  → Re-run E2E tests (max 3 iterations)
  → If still failing: ESCALATE to user

Step 4: Generate Output

## E2E Testing Summary
**Status:** {PASS|FAIL}
**Flows Tested:** {X/Y}
**Happy Path Tests:** {count}
**Error Path Tests:** {count}
**Browsers Passed:** {X/3}
**Consecutive Passes:** {X/3}

## Flow Coverage
| User Flow | Happy Path | Error Paths | Browsers | Viewports | Status |
|-----------|------------|-------------|----------|-----------|--------|
| {flow} | {PASS|FAIL} | {descriptions} | {X/3} | {X/3} | {PASS|FAIL} |

## Backend Handoff Verification
| Endpoint | Method | Contract Verified | Status |
|----------|--------|-------------------|--------|
| {endpoint} | {method} | {fields} | {PASS|FAIL} |

## Handoff to Next Gate
- Ready for Gate 6 (Performance Testing): {YES|NO}
- Iterations: {count}

Severity Calibration

Severity Criteria Examples
CRITICAL User flow completely broken, data corruption Transaction fails, user data lost, checkout blocked
HIGH Missing user flow tests, flaky tests, browser failure Untested flow from product-designer, test fails intermittently
MEDIUM Error path gaps, viewport issues Missing error handling test, mobile viewport untested
LOW Selector improvements, test organization Non-semantic selectors, test file structure

Report all severities. CRITICAL = immediate fix. HIGH = fix before gate pass. MEDIUM = fix in iteration. LOW = document.


Anti-Rationalization Table

See shared-patterns/shared-anti-rationalization.md for universal anti-rationalizations. Gate-specific:

Rationalization Why It's WRONG Required Action
"Unit tests cover the flow" Unit tests don't test real browser interaction. Write E2E tests
"Only Chromium matters" Firefox and Safari have different behavior. Test all 3 browsers
"Happy path is enough" Users encounter errors. Test error handling. Add error path tests
"CSS selectors are fine" CSS changes with refactors. Use semantic selectors. Use roles and testids
"Product-designer flows are suggestions" Flows define acceptance criteria. Cover all. Test all flows
"Test is flaky, skip it" Flaky tests indicate real instability. Fix the test