Claude Code Prometheus Monitoring Skill for RED Metrics
Connect request rate, errors and duration to real metrics and verify their Grafana panels using controlled traffic.
中文Install in Claude Code
Use a test service with a readable Prometheus and Grafana setup. Identify the actual metric names and labels first.
- Get the complete `plugins/devops/monitoring-stack-deployer/skills/deploying-monitoring-stacks` directory from the pinned GitHub checkout.
- Copy it under the target project's `.claude/skills/`. Confirm `.claude/skills/deploying-monitoring-stacks/SKILL.md`, open Claude Code at the project root, and call `/deploying-monitoring-stacks`.
Example task
Send a known mix of successful and controlled failed requests to the test service. Trace the metric behind each of three panels.What to check
The report identifies the rate, error and duration metrics and connects their observed changes to the test window.
Original SKILL.md on GitHub · Updated
Frequently asked question
How should I choose alert thresholds?
Use the service baseline, observation window and target, then test the rule against a controlled scenario.