skills.rest
witwave-ai
witwave-aiOFFICIAL
Explore all skills by this author →
WebsiteGitHub

platform-health

Inspects Kubernetes pods, operator status, and agent connectivity to report platform health.

2|Updated Apr 5, 2026
Search Tags:#monitoring#kubernetes#observability#health check#escalation#operator#pod restarts
One-click install
npx skills add https://github.com/witwave-ai/witwave --skill platform-health-witwave-ai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: platform-health
Source: https://github.com/witwave-ai/witwave/tree/main/.agents/self/mira/.claude/skills/platform-health
Command: npx skills add https://github.com/witwave-ai/witwave --skill platform-health-witwave-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Running a multi-agent platform on Kubernetes makes it hard to know whether pods, operators, storage, and inter-agent connectivity are actually healthy, and whether anomalies need escalation or are just normal churn. ## Core Features & Use Cases - Read-only health checks: Inspects pods, PVCs, operator status, releases, events, and resource metrics without mutating cluster or GitHub state. - Snapshot history and trend analysis: Appends compact JSON snapshots per run and compares restarts, token spend, and resource pressure across history to separate noise from systemic issues. - Escalation workflow: Produces a Green/Yellow/Red report and sends distilled anomaly handoffs to the Zora routing agent when a fix is needed. - Use Case: Ask whether it is safe to upgrade agents after noticing pod restarts; the skill checks restart deltas, OOM events, and operator health, then reports Yellow with evidence and a recommended handoff. ## Quick Start Run a platform health check on the witwave-self namespace and tell me if anything needs escalation.

Frequently Asked Questions about platform-health

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check the health of agents running on Kubernetes?▼

Run the platform health check, which inspects pod readiness, restart counts, operator status, and agent activity using kubectl and the ww CLI. It returns a Green, Yellow, or Red status with evidence-backed findings.

How to investigate pod restarts in a Kubernetes namespace?▼

The skill captures per-container restart counts, compares deltas against previous snapshots, and inspects the restarted pod's describe output, events, and previous container logs. It treats OOMKilled and repeated non-zero exits as escalation-worthy findings.

Can this skill upgrade or restart Kubernetes agents automatically?▼

No. The default posture is strictly read-only. Mutations like kubectl delete, rollout restarts, Helm upgrades, or ww operator upgrade require explicit human approval in the triggering request before any action is taken.

Does the health check work without a metrics server?▼

Yes. When the metrics API is unavailable, the skill records metrics_api_available as false and continues with the remaining checks. PVC capacity and resource ratios are marked null so historical comparison still works.

What happens when a systemic platform issue is detected?▼

The finding is recorded in an anomalies or escalations log, and a distilled evidence-backed handoff is sent to the Zora routing agent. The skill never performs the fix itself or dictates which mutation Zora should run.

Related Skills

Explore More Software Engineering Skills→
BjzyLabsBjzyLabs
1

kubectl

Manage Kubernetes clusters with kubectl using context-aware workflows.

Official
Intermediate
FlintmereFlintmere

security-claim-audit

Audits all of Flintmere's security claims across marketing, UI, docs, API, errors, and legal pages.

Official
Advanced
PostHogPostHog
713

investigating-metric-anomalies

Diagnoses server metric anomalies by correlating metrics, logs, and traces to root causes.

Official
Advanced
curiositechcuriositech
2

monitoring-stack-deployer

Deploy Prometheus, Grafana, and SLO-driven alerting on Kubernetes clusters.

Official
Advanced