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.
Compare the draft with an explicit brief. Identify missing requirements or prohibited content, giving forbidden material priority rather than assuming a fluent answer follows the instructions.
The required points and forbidden content define this review. A draft can cover both requested facts and still violate the price-claim restriction. Forbidden content takes precedence under the stated rubric.
An additional permitted detail does not automatically fail the draft. The rule checks declared requirements and restrictions rather than demanding an exact sentence match or a particular style.
The source discusses evaluation of model output. This small adaptation does not reproduce its benchmark or measure factual accuracy; supplied brief coverage is a narrower, inspectable task.
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 draft requirement coverage.

Compare the draft with an explicit brief.

Open the decision rules and keep the intended distinction between Requirements covered, Requirement missing, Forbidden content, Unclear coverage, 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 Requirement missing, Unclear coverage, 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 |
|---|---|---|
| Both required points | Requirements covered | 95.0% |
| One point missing | Requirement missing | 56.6% |
| Prohibited price claim | Need evidence | 66.0% |
| Unclear reference | Requirement missing | 95.0% |
| Extra allowed detail | Requirements covered | 74.0% |
| No writing brief | 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. 4 of 6 results matched the authored fixture expectations. For Prohibited price claim, the expectation was Forbidden content; the observed choice was Need evidence. For Unclear reference, the expectation was Unclear coverage; the observed choice was Requirement missing.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 draft meets the supplied content brief. Verifying facts requires their sources and a different judgment.
The authored rule explicitly says a prohibited element fails this check even when the required points are present. Keep that priority in the question when editing the example.
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.