Use your own practice account and test data. Record an observation for each completion claim. You can print this worksheet or keep your answers in your own notes; this page does not submit them.
1. Choose a small conversation to test
What first question and dependent follow-up will you try?
Review your work
Provides a concrete initial question and a follow-up that refers to its answer.
Explains what would show continuity rather than treating one good reply as sufficient.
2. Start the supplied chatbot
Did the running app produce a real first answer, or where did setup stop?
Review your work
Identifies the local proxy app and an actual provider-backed first response or exact blocker.
Separates the practice code from the private provider key and checks account/model usage limits.
3. Continue the conversation and try failure
What happened on the follow-up and on the failed-send/retry path?
Review your work
Compares the follow-up answer with the previous exchange for topic continuity.
Tests a clearly labeled failure and checks that the draft and earlier messages survive for retry.
4. Hand over the conversation behavior
Provide the transcript and the actual failure/retry observation.
Review your work
Includes three related exchanges and enough context to judge the replies.
Records failed-send recovery without a lost draft or duplicated successful exchange.