메뉴
Verify implemented features work correctly by driving browser/UI testing with screenshots and programmatic assertions
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Use when implementing features, fixing bugs, refactoring code, optimizing performance, resolving deprecations, or any multi-step development task requiring planning, implementation, testing, and review. Orchestrates specialized agent teammates through research, architecture, coding, QA, code review, adversarial review, and documentation phases. Triggers on: "implement", "build", "fix bug", "refactor", "add feature", "develop this", "help me build", "add functionality", "optimize performance", "resolve deprecation", "systematic development". Do NOT trigger on: simple questions ("what does this code do?"), file searches ("where is the auth function?"), one-off commands ("run the tests"), code explanations, quick edits, or non-development tasks.
Multi-lens adversarial review with Skeptic, Architect, and Minimalist lenses and 7+ attack vectors
Auto-improve any agent prompt using Karpathy's autoresearch method with iterative test-measure-improve loops
Safety guardrail that blocks destructive commands for the current session
Core development rules and philosophy for coding standards, git practices, and quality guidelines
Directory freeze guardrail that restricts file edits to a specific directory for the current session
| name | verify |
| description | Verify implemented features work correctly by driving browser/UI testing with screenshots and programmatic assertions |
Drive interactive testing to verify that implemented features actually work. Take screenshots, record evidence, and enforce programmatic assertions at each step. Uses Playwright MCP when available, falls back to CLI-based verification.
Triggers on: "verify this works", "check the output", "test the UI", "verify the feature", "screenshot test", "visual verification", "does it actually work". Do NOT trigger on: unit test writing (use tdd-workflow), code review (use code-review), standard test runs.
Announce: "Running verification. I will test the feature interactively and capture evidence."
Check for Playwright MCP (`mcp__playwright__*`) or Chrome DevTools MCP (`mcp__chrome-devtools__*`). Fall back to CLI verification if unavailable. From requirements/BDD scenarios, determine what to verify. Each step needs: action, expected result, assertion type. Navigate to feature, interact with UI, capture screenshots at each step, assert expected outcomes. Save screenshots, network logs, console output, assertion results. Pass/fail for each step with evidence links. Every assertion must be programmatic (not visual-only) — check element existence, text content, network responses Capture screenshot BEFORE and AFTER each interaction for comparison Report failures with: step number, expected vs actual, screenshot evidence If Playwright MCP unavailable, verify via CLI commands, API calls, or log output