Source and example scope
Compare each saved result with its authored expectation. A disagreement remains in the fixture review; the call is not repeated to obtain a preferred label.
Give the ad's specific promise and landing-page excerpt. Compare support, contradiction and a page that leaves the promise unestablished before revising the copy.
Promise not established preserves a gap in the supplied excerpt. It differs from an explicit statement that a feature is unsupported. Neither label requires guessing the rest of the page.
The broad-format fixture checks a universal promise against a narrow capability statement. A specific limitation is relevant evidence even when the page also lists a working feature.
This review compares supplied text only. It does not visit a live landing page, confirm actual product behavior or establish compliance with advertising rules.
These fictional inputs and decision rules were written for this page. The linked X discussion provides background; the technical implementation follows TypeSafe's Choice contract.

Use the sample to understand the rule, then test your own input for ad and landing-page alignment.

Give the ad's specific promise and landing-page excerpt.

Open the decision rules and keep the intended distinction between Promise supported, Promise contradicted, Promise not established, Unclear promise, Need evidence. Edit the choices only when testing your own rule.

Use a sample first, then try your own input. Compare the selected label with the other returned probabilities.

Review Unclear promise, Need evidence before using a result. The label is a decision about the supplied information, not an executed action.
| Original input | Observed label | Leading probability |
|---|---|---|
| Matching format support | Promise supported | 74.0% |
| Format conflict | Promise contradicted | 100% |
| Page stays silent | Promise not established | 94.0% |
| Vague promotional phrase | Unclear promise | 73.0% |
| Narrow support does not cover more | Promise contradicted | 100% |
| Page absent | Need evidence | 100% |
These are six original synthetic inputs, tested once per request revision. The percentage is the leading option probability, not measured accuracy. Authored expectations and any disagreements remain in the downloadable evidence. 6 of 6 results matched the authored fixture expectations.Verified
Compare each saved result with its authored expectation. A disagreement remains in the fixture review; the call is not repeated to obtain a preferred label.
The single-request SDK runner uses your private TYPESAFE_API_KEY. The offline consumer reads saved observations, checks request hashes and prints a pending output. It does not call the model or change any external system.
npm install @typesafe-ai/sdk@0.6.0
# Set TYPESAFE_API_KEY privately in your shell.
node run-choice-sample.mjs samples.json case-1
python preview_choice_workflow.py samples.json saved-api-results.jsonThe rubric explains the decision; the application owns the next step.
No. It means the supplied page text supports the promise. Product behavior needs independent testing.
Silence leaves the promise unestablished in this excerpt. Contradiction requires explicit conflicting evidence; preserving that distinction helps a reviewer request the right information.
No. It is the model distribution for this one request. The six examples are functional checks with authored expectations, not a representative benchmark. Inspect disagreements and evaluate your own examples before using the label.
The window shows an error and supplies no fresh judgment. The downloadable runner sends one request without retries. Need evidence is a valid outcome about incomplete input; it does not stand in for a network or provider failure.
Edit the evidence above or use the Playground to test another bounded question.