Build a useful investigation brief
Reconstruct the boundary
Record request, intent, context, retrieval results, tool state, prompt, response, confidence language, and expected answerability. Mark missing or conflicting evidence. Preserve versions and the exact response shown to the user.
Compare response choices
Test answer, qualified answer, clarifying question, useful partial response, and decline against the evidence. Identify unsupported claims, unnecessary refusal, hidden ambiguity, or a missing next step. Keep response length separate from evidence handling.
Test competing causes
Replay with complete, missing, conflicting, and clarified context. Check whether the response changes at the intended boundary. State whether the fault is evidence detection, intent interpretation, response policy, or evaluator ambiguity.
What to carry forward
The investigation brief should link request intent, evidence boundary, response choice, alternatives, replay, and user next action. End with a localized abstention error or evidence gap. Do not call a cautious answer wrong merely because it is shorter for the stated request and context under review.
Technical background: Google DeepMind evaluation research.
Keep the decision with the work.
Use a Work Item in Aglet to record the problem, the evidence you have, and the next decision. Add an owner and priority, then keep updates in the discussion so the next person can follow the reasoning.
Create an account See the product workflow