What problem does it solve? Modeling stateful processes like user lifecycles, issue triage, or retention loops in plain text is error-prone: states get orphaned, transitions lack labels, and diagrams break terminal width limits. This Skill produces rigorous, completeness-checked ASCII state machines that paste cleanly into PRs, ADRs, and growth docs. ## Core Features & Use Cases - Bracket-notation state machines: Renders [STATE] boxes with labeled --> transitions, guard conditions ([guard: x]), side-effect actions (/{action}), and terminal ((STATE)) markers. - Completeness verification: Checks every state has ingress and egress edges, every arrow has a trigger label, and recovery states have paths back to healthy states before delivery. - Metric annotation: Links each transition to a dashboard or query (e.g. metric: retention_rate_7d (grafana://dash/retention)) so the diagram stays falsifiable. - Mandatory validation gate: Every diagram passes bin/amw-validate-ascii.py and respects 78-column terminal / 100-column GitHub width ceilings. - Use Case: Ask for an ASCII state machine of your onboarding funnel, and receive a validated diagram of New → Activated → Retained → Churned with guards, metrics, and recovery edges ready to paste into a pull request. ## Quick Start Ask the AI to draw an ASCII state machine for your user lifecycle with states New, Activated, Retained, and Churned, including the trigger for each transition.