Shows what is broken in a Kubernetes cluster and why, and gives coding agents the same topology, events, changes and logs through a built-in MCP server. For platform and SRE teams.
- Access
- CLIMCPAPI
- Auth
- None
- Skills
- 5
Skills
Live issues
Ranks live operational failures: failing workloads, stuck Helm releases, dangling references and expiring certificates.
Diagnose
Collects a bounded evidence bundle for one workload, GitOps reconciler or network entry, with optional reachability probes.
Helm releases
Lists and inspects Helm releases with history, hooks, redacted values and revision diffs.
Prometheus queries
Runs PromQL against the cluster's Prometheus and discovers metric names and alerting rules.
AI investigations
Runs an investigation with your local Claude Code, Codex, Cursor or OpenCode CLI and ends on a findings page with evidence.
Related Agents
Kubernetes MCP Server
MCP · CLILets agents manage Kubernetes and OpenShift clusters through a native Go MCP server that talks to the API directly…
K8sGPT
CLI · API · MCPDiagnoses Kubernetes and OpenShift clusters in plain language, explaining why workloads are failing through a CLI…
Nightingale
MCP · APIEvaluates alert rules over metrics, logs, and SQL sources and routes notifications for SRE teams, with a built-in AI…
AgentOps
APIObservability platform purpose-built for AI agents. Track agent sessions, LLM calls, tool use, and costs. Integrates…