fold

Always active — base workflow framework for all Claude Code sessions. Defines phased development (research → plan → implement → verify → review → ship), tool selection, verification tiers, delegation patterns, and session conventions. Force-loaded via CLAUDE.md.

Security audit skill — two-track architecture: Track A (tool-based: static analysis → decide) and Track B (AI-based: context → review → decide). Wraps Trail of Bits plugins for language-agnostic security auditing. Commands: /security-audit static, /security-audit context, /security-audit review, /security-audit diff, /security-audit variants, /security-audit decide, /security-audit finding, /security-audit entry-points, /security-audit compliance, /security-audit cycle

Query and investigate Grafana observability data (Loki logs, Prometheus metrics, Tempo traces). Use this skill whenever the user mentions Grafana, asks about logs, metrics, traces, pod restarts, alerts, error rates, latency, OOM kills, crash loops, monitoring, observability, debugging production issues, or investigating any Kubernetes/infrastructure problem. Also triggers on alert messages pasted from Grafana or Slack (e.g. "[FIRING:N]"), or when the user wants to check dashboards, query Loki/Prometheus/Tempo, or correlate observability signals.

Manage Linear issues from the command line using the linear cli. This skill allows automating linear management.