observability-engineer — Claude Code skill: monitoring and SLOs
Walks you through monitoring, logging and tracing design, SLI/SLO targets and incident response. Useful when you run many services and alert noise gets out of hand.
npx claude-code-templates@latest --skill development/observability-engineerWhat observability-engineer does
The skill puts Claude in the role of a reliability engineer. It starts from the services and user journeys that matter, then moves to signals, instrumentation and retention, and only after that to dashboards and alerts tied to your SLOs. The last step is checking signal quality and cutting alert noise.
Stacks covered include Prometheus with PromQL and recording rules, Grafana, Loki, the ELK stack, OpenTelemetry, Jaeger, Kubernetes monitoring and cloud provider metrics. Separate sections handle error budgets, blameless postmortems, chaos experiments and observability as code.
There are clear boundaries too. A one-off dashboard is out of scope, and so is building application features. Without access to metrics, logs or traces the skill has nothing to work from. Secrets and sensitive data stay out of logs — the safety section says so directly.
What it helps with
- pick signals and instrumentation for a new service
- calculate an error budget and its burn rate
- cut alert noise and set up escalation
- trace a latency regression across services
- write a blameless postmortem procedure
Example prompts for the agent
- Define SLIs and SLOs for the payments service and work out the error budget
- Look through the traces and tell me why checkout latency went up
- Build a Grafana dashboard and alerts that match our SLOs
How to install observability-engineer
1claude-code-templates
npx claude-code-templates@latest --skill development/observability-engineer2Skills CLI (Vercel)
npx skills add davila7/claude-code-templates --skill observability-engineer -a claude-code3Manually: put the files in place
~/.claude/skills/observability-engineer/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/observability-engineer/
Manually: put the files in place
.agents/skills/observability-engineer/SKILL.mdFor all projects use ~/.agents/skills/
Manually: put the files in place
.cursor/skills/observability-engineer/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install davila7/claude-code-templates observability-engineer2Manually: put the files in place
.github/skills/observability-engineer/SKILL.mdPlatform CLI
gemini skills install https://github.com/davila7/claude-code-templates --path cli-tool/components/skills/development/observability-engineerManually: put the files in place
.devin/skills/observability-engineer/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about observability-engineer
How do I install observability-engineer in Claude Code?
Use the templates CLI: npx claude-code-templates@latest --skill development/observability-engineer. Or add it through skills: npx skills add davila7/claude-code-templates --skill observability-engineer -a claude-code.
What does observability-engineer do?
It works in four steps: identify critical services and reliability targets, define signals and retention, build dashboards and alerts around SLOs, then validate signal quality and reduce noise. Metrics, logs, tracing and incident handling all come into it.
When is this skill a poor fit?
When all you need is a single ad-hoc dashboard, or when the task is application feature work rather than observability. It also needs access to metrics, logs or tracing data to be of any use.
