What problem does it solve? Raw, untriaged GitHub issues filed by humans are often too vague for an autonomous development agent to pick up. This Skill converts such requests into well-formed, verifiable work items so the develop agent can act on them without ambiguity. ## Core Features & Use Cases - In-place issue shaping: Rewrites the existing GitHub issue body into ztrack form (summary plus 1-3 observable acceptance criteria) without changing its issue number. - Validation and labeling: Runs ztrack check on the drafted body, then marks the issue with the ready label and assigns it for dispatch. - Escalation handling: Labels vague requests needs-info and out-of-authority requests human-required instead of marking them ready. - Use Case: A maintainer files a one-line issue like "fix login timeout". The PM launches this Skill, which rewrites the issue with a source-grounded summary and provable acceptance criteria, validates it, and marks it ready for the develop agent. ## Quick Start Ask the agent to draft the raw GitHub issue referenced in $ZTRACK_ISSUE into a verifiable ready work item.