agentsclimarketplace

Validate feature qa

Skill lhenrique42/ai-templates/skills/validate-feature-qa

Validate feature implementation against PRD, technical spec, and task definitions using Playwright MCP E2E testing, WCAG 2.2 accessibility verification, and visual analysis. Documents bugs with screenshot evidence and produces a comprehensive QA report. Use when asked to run QA, validate a feature, or test implementation completeness. Do not use for code review, bug fixing, or task implementation.From its SKILL.md

Install
npx -y skills add lhenrique42/ai-templates --skill validate-feature-qa

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

5.3 KB, ~1.0k tokens by cl100k_base, as published. Nobody here has run it

Validate Feature QA

Outcome

Produce a comprehensive QA validation package for a feature implementation, including:

  • End-to-end test execution evidence (Playwright MCP)
  • WCAG 2.2 accessibility verification results
  • Visual analysis findings with screenshots
  • Bug records with reproducible steps and severity
  • A final QA report with readiness verdict

When to Use

  • User asks to run QA on a feature
  • User asks to validate implementation completeness
  • User asks to verify behavior against PRD, tech spec, and tasks

Do Not Use

  • Performing code review
  • Implementing fixes or patching defects
  • Executing task implementation work

Required Inputs

  • Feature name or target area
  • PRD path/content
  • Tech spec path/content
  • Task artifacts (tasks.md and relevant task files)
  • Test environment URL or run instructions
  • Optional accounts/test data and browser/device constraints

Workflow

1) Resolve Scope and Acceptance Baseline

Extract what must be validated from PRD, tech spec, and tasks.

Capture:

  • Functional requirements (FR-*) and constraints
  • Non-functional expectations (NFR-*)
  • Task acceptance criteria and definition of done
  • Explicit out-of-scope boundaries

Decision points:

  • If any core artifact is missing (PRD, tech spec, or tasks), stop and request it.
  • If artifacts conflict, log ambiguity and request clarification before final verdict.

2) Build QA Coverage Matrix

Create a traceable checklist mapping requirements to tests.

Include:

  • Requirement ID -> test scenario(s)
  • Task ID -> validation status
  • NFR checks -> evidence method

Decision points:

  • If requirements cannot be mapped to executable checks, raise testability gap.
  • If tasks include incomplete dependencies, flag risk before running full suite.

3) Prepare Test Session

Prepare deterministic test preconditions.

Prepare:

  • Environment health/smoke check
  • Test accounts and seed data
  • Browser coverage targets (minimum agreed set)
  • Screenshot output location

Decision points:

  • If environment is unstable, record as blocker and continue only with partial validation if feasible.
  • If required data is unavailable, log blocked scenarios explicitly.

4) Run E2E Validation with Playwright MCP

Execute core user workflows and edge cases using Playwright MCP.

Run:

  • Happy-path scenarios
  • Negative and edge scenarios
  • Regression-critical paths touched by feature changes

Evidence rules:

  • Capture screenshots for failed checks and notable visual issues.
  • Record exact scenario name, step, and observed vs expected behavior.

5) Perform Accessibility Verification (WCAG 2.2)

Validate accessibility conformance for impacted screens and flows.

Check at minimum:

  • Keyboard navigation and focus order
  • Focus visibility and logical tab sequence
  • Form labels, names, roles, values
  • Color contrast and non-text contrast
  • Error identification and messaging
  • Reduced motion and interaction clarity where applicable

Decision points:

  • If severe accessibility blockers are found, mark feature as not ready.
  • If checks are partially blocked by environment limits, document the gap and residual risk.

6) Perform Visual Analysis

Verify visual correctness and consistency across target states.

Check:

  • Layout integrity and responsive behavior
  • Content overlap, clipping, or truncation
  • State-specific UI correctness (loading, empty, error, success)
  • Cross-browser visual anomalies (if in scope)

Evidence rules:

  • Attach screenshot evidence for each visual defect.
  • Tag each issue with impacted viewport/browser context.

7) Document Bugs with Evidence

Create bug entries using Bug report template.

For each bug capture:

  • Severity (Blocker, Critical, Major, Minor)
  • Requirement/task traceability
  • Reproduction steps
  • Expected vs actual behavior
  • Screenshot path(s)

8) Generate Comprehensive QA Report

Create final report using QA report template.

Report must include:

  • Coverage matrix summary
  • Pass/fail status by requirement and task
  • Accessibility and visual findings
  • Bug summary by severity
  • Readiness verdict (Pass, Pass with Known Issues, Fail, Blocked)

Quality Check Before Finalizing

QA validation is complete only if:

  • Every in-scope requirement/task has a mapped validation status.
  • All discovered bugs are documented with reproducible steps.
  • Every bug includes screenshot evidence where applicable.
  • Accessibility checks explicitly reference WCAG 2.2 criteria.
  • Residual risks and blocked checks are clearly listed.
  • Final readiness verdict is justified by evidence.

Output Contract

Return a structured QA report in chat and optionally save markdown artifacts for the QA report and bug list using:

What ships with it: 2 files

2.3 KB alongside SKILL.md

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.