Rollback: undoing what the agent did

Your first line in any agent design is what the agent must not do. Most failures come from unclear scope, sloppy tool contracts and missing failure handling, not from a weak model.

Task: The agent changed config or sent messages and it must be undone.

[Before you start, do three things]
1. Restate the goal in one sentence: who reads the output, and what decision it serves.
2. If I have not given you the following, ask me. Do not assume:
   - which actions are reversible
   - the rollback window
   - have dependents consumed the change
   - who can trigger it
3. Say how you plan to do it (at most 5 steps) and wait for my go-ahead.

[Go through each of these]
- save prior state before any reversible action
- list irreversible actions and their remedies
- the rollback must be idempotent and verifiable
- notify affected dependents after rollback

[Process]
1. Write the boundary: what it may do, must never do, and when to hand back to a human
2. Define each tool's inputs, outputs, failure semantics, idempotency
3. Write the flow as a state machine with exit conditions
4. Design failure paths: retry, degrade, abort, escalate
5. Give one testable success criterion

[Output]
Fixed shape: boundary, tool contracts, state machine, failure paths, criteria.

[Do not]
- No general-purpose agent that does everything
- Never let fuzzy prose stand in for a tool contract
- The agent must not decide whether approval is needed
- No 'do your best' instead of an abort condition