What problem does it solve? Executing a multi-phase implementation plan by hand requires constant human supervision, manual test runs, and ad-hoc commit decisions. This Skill runs an approved plan from context/changes/<change-id>/plan.md end-to-end without any human interaction, enforcing quality gates and producing an auditable commit trail. ## Core Features & Use Cases - Autonomous phase execution: Reads the plan's ## Progress section as the single source of truth, implements each phase, and flips Automated checkboxes as work completes. - Fixed quality-gate stack: Runs plan success criteria, a deliberate-break test check, and repo-wide lint/typecheck/test suites before every commit, with a two-attempt self-fix budget. - Conventional Commits ritual: Commits each green phase with a structured message, writes the short SHA back into Progress rows, and finishes with an epilogue commit and run report. - Use Case: Point it at a planned change in a headless session (e.g. your AI coding assistant -p with a /goal condition) and return later to find all phases implemented, committed, and a human checklist of pending Manual verification rows. ## Quick Start Run /goal with a completion condition followed by /10x-goal-implement <change-id> to autonomously implement every phase of that change's plan.
