菜单
Integrates ProofShot visual proof recording with the self-improvement workflow. Use when: (1) E2E tests fail or detect visual regressions, (2) Agent needs to verify UI changes it made, (3) Running self-iteration loops (build → test → fix → retest), (4) Preparing PR proof artifacts for review.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | e2e-self-iteration |
| description | Integrates ProofShot visual proof recording with the self-improvement workflow. Use when: (1) E2E tests fail or detect visual regressions, (2) Agent needs to verify UI changes it made, (3) Running self-iteration loops (build → test → fix → retest), (4) Preparing PR proof artifacts for review. |
| metadata | null |
Combines ProofShot (visual proof recording) and agent-browser (headless browser control) with the self-improvement loop for automated E2E testing, visual regression detection, and iterative bug fixing.
| Situation | Action |
|---|---|
| Need to verify UI changes | vx just e2e-proofshot |
| Found visual regression | Log to .learnings/ERRORS.md, fix, re-run |
| Self-iteration loop | vx just e2e-iterate |
| PR needs proof | vx just e2e-pr |
| Explore UI elements | vx just e2e-snapshot |
| Compare against baseline | vx just e2e-diff |
# Install ProofShot + agent-browser
vx just e2e-install
After making UI changes, verify them visually:
vx just e2e-start # Build + start Gallery with CDP
vx just e2e-snapshot # See interactive elements
vx just e2e-screenshot # Capture annotated screenshot
vx just e2e-stop # Cleanup
Record a complete verification session with video + artifacts:
vx just e2e-proofshot # Full automated flow
This will:
When fixing bugs or implementing features, iterate:
# Cycle: build → start → verify → fix → repeat
vx just e2e-iterate
# Review artifacts after each iteration
ls proofshot-artifacts/
Direct browser control commands:
# Element discovery
vx just e2e-snapshot
# Navigation
vx just e2e-open "http://localhost:5173/settings"
# Interaction
vx just e2e-click "@e5"
vx just e2e-fill "@e2" "test value"
# Screenshot
vx just e2e-screenshot
Compare current state against baseline:
vx just e2e-diff
Attach visual proof to a Pull Request:
vx just e2e-pr # Auto-detect PR
vx just e2e-pr 42 # Specific PR number
When the agent detects issues during E2E testing:
.learnings/ERRORS.md
## [ERR-YYYYMMDD-XXX] e2e_visual_regression
**Logged**: ISO-8601 timestamp
**Priority**: high
**Status**: pending
**Area**: frontend
### Summary
[Description of what's wrong visually]
### Error
[Console errors if any]
### Context
- Screenshot: proofshot-artifacts/<session>/step-*.png
- Page: [URL or route]
- Expected: [what it should look like]
- Actual: [what it looks like]
### Suggested Fix
[Specific code change needed]
vx just e2e-iterate againresolved in .learnings/ERRORS.md# CI pipeline step
vx just e2e-ci
# Upload proof to PR after CI tests
vx just e2e-pr
proofshot-artifacts/
└── <timestamp>/
├── session.webm # Full session video
├── viewer.html # Interactive viewer
├── SUMMARY.md # Markdown report
├── step-*.png # Key screenshots
├── session-log.json # Action timeline
├── server.log # Server stdout/stderr
└── console-output.log # Browser console
Convert an existing Qt/PySide/PyQt desktop project (QWebEngineView-based UI, QMainWindow browsers, Qt DCC tools) into an AuroraView project so it ships a lighter, Rust-powered WebView and gets full MCP automation for free. Use this skill whenever the user asks to "migrate", "convert", "port" a Qt/PySide/PyQt app to AuroraView, or when they want MCP-controllable Qt tooling.
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.
This skill teaches AI coding agents how to use vx - the universal development tool manager. Use this skill whenever the user's project uses vx (has vx.toml or .vx/ directory), or when the user mentions vx, tool version management, or cross-platform development setup. vx transparently manages Node.js, Python, Go, Rust, and 50+ other tools with zero-config.
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
This skill produces an actionable architecture diagnosis report for a codebase. It guides the AI to scan the repository, map subsystem boundaries, identify design flaws with concrete evidence (files/functions), assess risks, and propose a phased refactor roadmap plus future extension scenarios. Use this skill when the team needs a "diagnosis report" to decide whether/how to refactor.
通过 MCP 工具测试 AuroraView Python API，发现设计缺陷和代码味道。 迭代式测试：启动服务 → 调用工具 → 发现问题 → 修复 → 重新测试。