Build a useful investigation brief
Build crossed configurations
Create matching test and production fixtures plus a crossed endpoint and credential case. Label expected destination and safe outcome before execution. Include a copied resource identifier and a response without an environment marker only if the client has a defined review state.
Trace selection to response
Capture configuration source, selected base URL, credential label, outbound endpoint, response marker, resource identifier, and local state. Compare the first point where expected and actual environment diverge. Redact secrets and avoid using real production records in the fixture.
Challenge the default
Vary one variable, override, endpoint, identifier, or process restart at a time. Compare whether the mixup follows configuration or resource lookup. If the source of an environment marker is remote and unobservable, document that boundary rather than infer it.
What to carry forward
The investigation is ready when matching and crossed fixtures show endpoint, credential, resource, and local-state transitions. Deliver a narrow configuration or labeling change and any record review. Leave unmarked remote resources explicitly unknown for review.
Technical background: OpenAI documentation.
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