All categories/Agents and automation
Debugging agents · Level Advanced · Debugging Agent Hallucination
Hallucination: why did it invent a number
The agent cited a number or source that does not exist.
Prompt
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: The agent cited a number or source that does not exist.
[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:
- the answer and question
- sources available
- citation required
3. Say how you plan to do it (at most 5 steps) and wait for my go-ahead.
[Go through each of these]
- decide whether sources were missing or misread
- mark every sentence needing external support
- constrain: if no evidence, say it was not found
- when citing, give the location so it can be checked
[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 casesWhat to state before it starts
Leave these unsaid and the model fills them with an assumption. Most 'wrong answers' start here.
- the answer and question
- sources available
- citation required
Go through each of these
Where this kind of work most often goes wrong. Already in the prompt, so you do not have to remember it.
- decide whether sources were missing or misread
- mark every sentence needing external support
- constrain: if no evidence, say it was not found
- when citing, give the location so it can be checked