Cost control: what one task costs

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: Token spend runs away and the bill is alarming.

[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:
   - token distribution per task
   - the most expensive steps
   - acceptable unit cost
   - caching or smaller models
3. Say how you plan to do it (at most 5 steps) and wait for my go-ahead.

[Go through each of these]
- break token use down by step and name the top three
- propose context compression: summaries, retrieval instead of full text
- move simple classification to a smaller model
- set a per-task budget and what happens on breach

[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