Agent Skills: plan-execute

This skill should be used for any non-trivial request — features, bugs, stories, epics, spikes, or multi-step tasks. It accepts a ticket URL (Jira, Linear, GitHub), a file path containing a spec, or a plain-text prompt. It assembles an agent team, breaks the work into structured tasks, and manages the full lifecycle from research through implementation, code review, deploy, and empirical verification.