Stop Asking and Just Do It

This skill is coherent with its stated goal of doing code work autonomously, but it purposely instructs the agent to act without asking for confirmation and to avoid clarifying questions. Before installing or enabling it, consider whether you want the agent to: (1) make unilateral edits to your repository or create commits/PRs without explicit approval, (2) proceed on assumptions that might be wrong for your project, or (3) perform risky refactors or wire-ups that should be reviewed. If you still want autonomy but with limits, ask the author (or modify the SKILL.md) to add safe-guards: require explicit confirmation before writing to the repo or pushing, produce a proposed plan/patch for review before applying it, never push to remote without approval, and log a clear summary of all changes. If you cannot inspect or control those behaviors, treat this skill as high-risk and avoid enabling it for repositories where unintended changes are unacceptable.