Manage Grafana Cloud resources through the gcx CLI with kubectl-style commands.
npx skills add https://github.com/titaneric/dotfiles --skill gcx-titaneric
Please help me install this Agent Skill. Skill: gcx Source: https://github.com/titaneric/dotfiles/tree/main/dot_agents/skills/gcx Command: npx skills add https://github.com/titaneric/dotfiles --skill gcx-titaneric
gcx resources and dedicated command groups. - Signal Queries: Run PromQL, LogQL, Pyroscope, and Tempo trace queries directly, with LLM-friendly compact output for trace inspection. - Safe Mutation Workflow: Verify context, read current state, build from schemas/examples, preview with --dry-run, apply, and verify changes. - Grafana Assistant Integration: Use gcx assistant prompt and investigations for cross-signal root cause analysis when deterministic queries are not enough. - Use Case: An on-call engineer sees a firing alert, runs gcx alert instances list --state firing, queries related traces with gcx traces query, then asks the Assistant why the latency spiked — all without leaving the terminal. ## Quick Start Ask the AI to list all firing alert instances in your current Grafana Cloud context using gcx, then investigate the root cause.High-intent search queries and answers about installing and using this skill.
Use the gcx resources group: `gcx resources get dashboards` to list, `gcx resources push` to create or update from local files, and `gcx resources pull` to export. Discover schemas with `gcx resources schemas dashboards` before building payloads.
Use `gcx metrics query -d <datasource-uid> 'up'` for PromQL and `gcx logs query -d <uid> '{app="foo"}'` for LogQL. These signal-specific commands handle datasource resolution automatically and are preferred over raw datasource queries.
Follow the safe mutation workflow: verify the active context with `gcx config check`, read current state, build payloads from `gcx resources schemas` and `examples`, preview with `--dry-run`, apply, then re-read the resource to confirm the change.
Use deterministic commands when you know the exact query, since they are faster and cheaper. Use `gcx assistant prompt` or investigations when you need cross-signal root cause analysis or do not know which metrics and labels exist.
Yes, gcx uses named contexts like kubectl. Run `gcx config current-context` to see the active one, `gcx config use-context <name>` to switch, or pass `--context <name>` on any command to target a specific environment without switching.
pytorchOperates and troubleshoots OSDC monitoring and logging pipelines in Kubernetes and Terraform environments.
AbdullahMalik17Define SLOs, error budgets, and blameless postmortems for production service reliability.
SigNozCorrelate SigNoz alert signals with metrics, traces, and logs to diagnose root causes.
slopusQuery and manage Grafana dashboards and Prometheus metrics via grafanactl CLI.