Monitors deployed URLs post-release for HTTP status, SSE streams, static assets, console errors, and performance regressions. Use for smoke tests, canary checks, and deployment validation after deploys, risky merges, or dependency upgrades.
How this skill is triggered — by the user, by Claude, or both
Slash command
/everything-claude-code:canary-watchThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
- 部署到生产环境或预发布环境后
监控已部署 URL 是否出现退化。循环运行直到被停止或监控窗口到期。
1. HTTP 状态 — 页面是否返回 200?
2. 控制台错误 — 是否有之前没有的新错误?
3. 网络失败 — API 调用失败、5xx 响应?
4. 性能 — LCP/CLS/INP 相对于基线是否退化?
5. 内容 — 关键元素是否消失了?(h1、导航、页脚、CTA)
6. API 健康 — 关键端点是否在 SLA 内响应?
7. 静态资源 — JS、CSS、图片和字体请求是否返回 2xx/3xx 且内容类型符合预期?
8. SSE 流 — 事件流端点是否能连接并接收初始事件或心跳?
快速检查(默认):单次通过,报告结果
/canary-watch https://myapp.com
持续监控:每 N 分钟检查一次,持续 M 小时
/canary-watch https://myapp.com --interval 5m --duration 2h
对比模式:对比预发布环境与生产环境
/canary-watch --compare https://staging.myapp.com https://myapp.com
critical: # 立即告警
- HTTP 状态 != 200
- 控制台错误数 > 5(仅新错误)
- LCP > 4s
- API 端点返回 5xx
- 静态资源返回 4xx/5xx
- SSE 端点无法连接或在首次心跳前断开
warning: # 在报告中标记
- LCP 相对基线增加 > 500ms
- CLS > 0.1
- 新的控制台警告
- 响应时间 > 基线的 2 倍
- 静态资源内容类型意外改变
- SSE 心跳延迟 > 基线的 2 倍
info: # 仅记录
- 轻微性能波动
- 新的网络请求(添加了第三方脚本?)
当超过严重阈值时:
~/.claude/canary-watch.log## 金丝雀报告 — myapp.com — 2026-03-23 03:15 PST
### 状态:健康 ✓
| 检查项 | 结果 | 基线 | 差异 |
|-------|------|------|------|
| HTTP | 200 ✓ | 200 | — |
| 控制台错误 | 0 ✓ | 0 | — |
| LCP | 1.8s ✓ | 1.6s | +200ms |
| CLS | 0.01 ✓ | 0.01 | — |
| API /health | 145ms ✓ | 120ms | +25ms |
| 静态资源 | 42/42 ✓ | 42/42 | — |
| SSE /events | 已连接 ✓ | 已连接 | +80ms 心跳 |
### 未检测到退化。部署正常。
配合使用:
/browser-qa 用于部署前验证git push 的 PostToolUse 钩子,在部署后自动检查npx claudepluginhub aaione/everything-claude-code-zhMonitors deployed URLs after releases: checks HTTP status, console errors, static assets, SSE streams, and performance regressions. Use after deploys, risky merges, or dependency upgrades.
Monitors deployed web app URLs for post-deploy regressions including HTTP status, console errors, network failures, performance metrics (LCP/CLS/INP), content changes, and API health with alerts and looping checks.
Verifies production site health post-deploy via canary monitoring: checks HTTP status, response time, and error patterns, compares against baseline to detect regressions, and guides rollback decisions.