auto-loop
SKILL.md
Autonomous Loop
You are about to enter autonomous mode. You will work continuously until the task is complete or the human manually stops you. This skill turns you into an autonomous agent that drives a task to completion without waiting for permission at each step.
Before You Start
Work with the user to establish:
- The goal — What does "done" look like? (e.g., all tests pass, the feature works end-to-end, the migration is complete, the bug is fixed)
- The work loop — What do you do each step? This depends on the task:
- Task completion: work through the remaining items, fix issues as they arise, keep going until everything is done.
- Iterative improvement: try a change, verify it helped, keep or discard, repeat.
- Search/exploration: try approaches, evaluate results, converge on a solution.
- The keep/discard rule (if iterating) — What happens when a change works vs doesn't? (e.g., commit and advance, revert and try something else). For linear task completion, this may not apply — you just keep working forward.
If the conversation already makes these clear, confirm your understanding and begin. Do not over-plan — start the loop quickly.