跳至正文
OpenAgentSkill
浏览已开放 · 目录记录日期 2026-11-21 · 自动更新和账户服务暂时关闭。
部署与云服务

monitoring

Workflow for implementing comprehensive monitoring and alerting. Use when the user needs to set up or improve service monitoring.

girijashankarj · 已有静态检查记录

安装与使用

运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。

npx skills add girijashankarj/cursor-handbook --skill monitoring

技能说明


name: monitoring description: Workflow for implementing comprehensive monitoring and alerting. Use when the user needs to set up or improve service monitoring.

Skill: Set Up Monitoring

Trigger

When the user needs to set up or improve service monitoring.

Steps

Step 1: Define Metrics

  • Request rate (throughput)
  • Error rate (4xx, 5xx)
  • Latency (p50, p95, p99)
  • Resource utilization (CPU, memory, disk)
  • Business metrics (domain-specific)

Step 2: Implement Structured Logging

  • JSON format for all logs
  • Required fields: timestamp, level, correlationId, service
  • No PII in logs
  • Appropriate log levels (error, warn, info, debug)

Step 3: Set Up Tracing

  • Add OpenTelemetry or provider-specific SDK
  • Instrument HTTP requests
  • Instrument database queries
  • Instrument external service calls
  • Configure sampling rate

Step 4: Create Dashboards

  • Service health overview
  • Request rate and error rate
  • Latency percentiles
  • Resource utilization
  • Business KPIs

Step 5: Configure Alerts

  • Error rate > 1% → P3
  • Error rate > 5% → P1
  • p99 latency > 2s → P3
  • CPU > 80% → P3
  • Health check failure → P1
  • Set notification channels

Step 6: Create Runbooks

  • For each alert, write a runbook:
    • What the alert means
    • Investigation steps
    • Resolution steps
    • Escalation path

Completion

Monitoring, alerting, and runbooks are in place.