Enforces end-user perspective validation through real system execution. Triggers on "validate", "functional test", "prove it works", "evidence". Never write mocks or test files. Validate via simulator, browser, CLI, or cURL with PASS/FAIL verdict.
From cc-setupnpx claudepluginhub krzemienski/cc-setup --plugin cc-setupThis skill uses the workspace's default tool permissions.
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.
Details PluginEval's skill quality evaluation: 3 layers (static, LLM judge), 10 dimensions, rubrics, formulas, anti-patterns, badges. Use to interpret scores, improve triggering, calibrate thresholds.
NEVER create mocks, stubs, test doubles, or test files.
NEVER import jest, vitest, pytest, XCTest, or any test framework.
ALWAYS validate through the same interfaces real users experience.
ALWAYS capture evidence that proves the feature works end-to-end.
Detect the platform FIRST. Wrong platform = wrong validation approach.
| Priority | Platform | Indicators | Approach |
|---|---|---|---|
| 1 | iOS/macOS | *.xcodeproj, Package.swift | Xcode build, simulator, simctl/idb |
| 2 | Web | package.json + React/Vue/Next | Dev server, browser, Playwright MCP |
| 3 | CLI | main.go, Cargo.toml, cli.py | Build binary, execute, capture stdout |
| 4 | API | server.ts, app.py + routes | Start server, curl endpoints |
| 5 | Full-Stack | Frontend + Backend present | Bottom-up: DB -> API -> Frontend |
Build the real system with all real dependencies. If the build fails, that IS your first finding — stop and report it before proceeding.
Interact as a real user would — browser, simulator, CLI binary, or curl. No REPL imports, no direct function calls, no internal API invocations that bypass the real entry point.
Save artifacts to e2e-evidence/ — screenshots, response bodies, CLI output, logs. Evidence must be READ and DESCRIBED with specific quoted content, not just confirmed to exist.
For each criterion, cite specific evidence with file paths and quoted content. See verdict format below.
For full-stack apps, validate bottom-up. Each layer must PASS before testing the layer above it.
Database/Infra -> Business Logic -> API Endpoints -> Frontend UI
(first) (last)
If the DB is broken, every API test fails with misleading errors. Fix lower layers first.
### Criterion: [What was required]
**PASS** / **FAIL**
Evidence: `e2e-evidence/[file]` — [What I actually saw, quoted specifically]
All criteria must be evaluated before claiming completion. A single FAIL blocks the completion claim until fixed and re-validated.
This skill executes real systems for validation. It never introduces new functionality, never disables security checks, and never bypasses auth — it validates through the same security boundaries real users encounter.