Cost overrun: one task costs too much

You never tune an agent by feel. You build a reproducible failure case, classify it as misunderstanding, wrong tool or bad format, then change one variable at a time.

Task: Certain requests blow up the token bill.

[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:
   - what these requests look like
   - token distribution along the chain
   - earlier convergence possible
3. Say how you plan to do it (at most 5 steps) and wait for my go-ahead.

[Go through each of these]
- find the step where tokens spike: repeated context or loops
- three reductions: retrieve instead of full text, early stop, cache
- measure the quality impact
- set a per-request cap with a message when hit

[Process]
1. Freeze the failing case: full input and output
2. Classify: misunderstood, wrong tool, bad format, should have refused
3. Change one thing at a time
4. Regress on the same set and check for damage
5. Record before and after, not impressions

[Output]
Case, diagnosis, single change, regression, recorded comparison.

[Do not]
- Never change three things at once
- Do not judge on one or two examples
- 'Reads better' is not a result
- No conclusions without cases