# Kubeblocks Observability Router

> Primary observability entry for KubeBlocks-managed databases. Use this when the user broadly asks for monitoring or observability and you need to choose between integrating with an existing stack, bootstrapping a new stack, or declaring the current readiness level. The legacy kubeblocks-setup-monitoring skill remains only as a compatibility shim.

- Skill: `apecloud/kubeblocks-observability-router` (Agent Skill)
- Install (CLI): `npx skillmds@latest add apecloud/kubeblocks-observability-router`
- Raw SKILL.md: https://api.skillmd.com/api/skills/apecloud/kubeblocks-observability-router/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: apecloud (https://skillmd.com/u/apecloud)
- Updated: 2026-09-10
- Page: https://skillmd.com/skills/apecloud/kubeblocks-observability-router

---


# Observability Router

Use [observability-capability-matrix](../../references/coverage/observability-capability-matrix.yaml) as the machine-readable truth for exporter evidence, scrape evidence, dashboard evidence, alerting evidence, and the maximum readiness ceiling for each Tier-1 engine.

## Routing Contract

- If Prometheus/Grafana already exists, route to [kubeblocks-observability-existing-stack](../kubeblocks-observability-existing-stack/SKILL.md).
- If no monitoring base exists but the engine has exporter evidence, route to [kubeblocks-observability-bootstrap-stack](../kubeblocks-observability-bootstrap-stack/SKILL.md).
- If the matrix says `exporter: false`, do not claim `scrape-ready` or higher without fresh evidence from the target environment.
- Always state whether the current state is `none`, `metrics-ready`, `scrape-ready`, `dashboard-ready`, or `alerting-ready`.
- If observability evidence is unclear or the live cluster state disagrees with the matrix, route to [kubeblocks-troubleshoot](../kubeblocks-troubleshoot/SKILL.md) before promising a higher readiness level.