This handles the full chaos engineering workflow, from mapping dependencies and designing experiments to executing controlled failures and writing post-mortems. You get experiment manifests for Litmus Chaos and Chaos Monkey, toxiproxy network injection scripts, safety checklists that enforce blast radius caps and sub-30-second rollback paths, and game day runbooks. The safety constraints are sensible: single variable changes, steady state verification first, no production without circuit breakers. Useful if you're moving past ad-hoc failure testing and want structured experiment design with actual rollback procedures, not just "delete a pod and see what happens."
npx skills add https://github.com/jeffallan/claude-skills --skill chaos-engineer