菜单
Operate Prometheus-compatible monitoring with kubectl, curl, PromQL, and VictoriaMetrics CLI/API endpoints. Use for metrics, active alerts, cluster health checks, resource utilization, performance bottlenecks, SLOs, and capacity planning. 适用于指标查询、告警分析、集群巡检、资源使用率、性能问题和容量规划。
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
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。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/clcc2019/agentic-infra --skill monitoring-ops命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | monitoring-ops |
| description | Operate Prometheus-compatible monitoring with kubectl, curl, PromQL, and VictoriaMetrics CLI/API endpoints. Use for metrics, active alerts, cluster health checks, resource utilization, performance bottlenecks, SLOs, and capacity planning. 适用于指标查询、告警分析、集群巡检、资源使用率、性能问题和容量规划。 |
Confirm the kubectl context, discover the in-cluster or configured query endpoint, then use curl --get --data-urlencode 'query=<promql>' <endpoint>/api/v1/query. Prefer bounded time ranges and record the query and timestamp with each conclusion.
# Node status
kube_node_status_condition{condition="Ready",status="true"}
# Pod status counts
count(kube_pod_status_phase{phase="Running"})
count(kube_pod_status_phase{phase="Pending"})
count(kube_pod_status_phase{phase="Failed"})
# Node CPU usage
100 - (avg by(instance)(rate(node_cpu_seconds_total{mode="idle"}[5m])) * 100)
# Node memory usage
(1 - node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes) * 100
# Pod CPU usage (based on Limit)
sum(rate(container_cpu_usage_seconds_total{container!=""}[5m])) by (pod, namespace)
/ sum(kube_pod_container_resource_limits{resource="cpu"}) by (pod, namespace) * 100
# Pod memory usage (based on Limit)
sum(container_memory_working_set_bytes{container!=""}) by (pod, namespace)
/ sum(kube_pod_container_resource_limits{resource="memory"}) by (pod, namespace) * 100
| Metric | Warning | Critical | Description |
|---|---|---|---|
| CPU usage | >70% | >85% | For 5 minutes |
| Memory usage | >75% | >90% | For 5 minutes |
| Disk usage | >80% | >90% | Single check |
| Pod restarts | >3 | >10 | Last 1 hour |
| PVC usage | >80% | >90% | Single check |
# API Server request latency P99
histogram_quantile(0.99, rate(apiserver_request_duration_seconds_bucket{verb!="WATCH"}[5m]))
# API Server error rate
sum(rate(apiserver_request_total{code=~"5.."}[5m])) / sum(rate(apiserver_request_total[5m])) * 100
# ETCD Leader changes
changes(etcd_server_is_leader[1h])
# ETCD database size
etcd_mvcc_db_total_size_in_bytes
# ETCD disk fsync latency
histogram_quantile(0.99, rate(etcd_disk_wal_fsync_duration_seconds_bucket[5m]))
# CPU trend forecast (7-day linear regression)
predict_linear(node_cpu_seconds_total{mode="idle"}[7d], 30*24*3600)
# Disk usage trend forecast
predict_linear(node_filesystem_avail_bytes[7d], 30*24*3600)
Keep alert-rule changes in Git, validate them with promtool check rules, show the diff, and require approval before apply or reload.