Agent Skills
› AjayIrkal23/agentic-mercy-10x
› canary-playwright
canary-playwright
GitHub用于部署后金丝雀监控，通过 Playwright MCP 执行健康检查。涵盖页面加载、认证流程、关键用户路径测试及 API/控制台错误检测，最终输出带证据的 PASS/FAIL 报告。
Trigger Scenarios
部署完成后监控
生产环境健康检查
Install
npx skills add AjayIrkal23/agentic-mercy-10x --skill canary-playwright -g -y
SKILL.md
Frontmatter
{
"name": "canary-playwright",
"schema": 1,
"category": "general",
"surfaces": [
"general"
],
"triggers": {
"paths": [],
"intents": [
"general"
],
"keywords": [
"adapted",
"canary",
"gstack",
"mcp",
"monitoring",
"playwright",
"post-deploy"
]
},
"platforms": [
"linux",
"darwin",
"windows"
],
"token-cost": 170,
"description": "ALWAYS invoke for post-deploy canary monitoring using Playwright MCP (adapted from GStack \/canary)."
}
Canary Monitoring
Post-deploy health checks using Playwright.
Steps
- Navigate to production/staging URL:
mcp__playwright__browser_navigate - Health check — verify page loads, no error screens
- Auth flow — login if applicable, verify session works
- Critical paths — test 2-3 most important user flows
- API health — check network requests for errors:
mcp__playwright__browser_network_requests - Console — check for new errors:
mcp__playwright__browser_console_messages - Screenshot final state:
mcp__playwright__browser_take_screenshot
Report
Output: PASS/FAIL with evidence (screenshots, console errors, failed requests).
Version History
- 581d130 Current 2026-07-19 09:05
