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.
Provide the requirements and submission notes. Compare required parts that are present, explicitly absent or only mentioned before passing the packet to a reviewer.
Reference unresolved keeps a mentioned but unconfirmed reviewer note separate from an explicitly missing note. That distinction tells an intake reviewer whether to verify an attachment or request a new document.
The optional-budget fixture checks the actual checklist. An absent optional part does not become a missing requirement because its name appears in the submission notes.
This text review does not open attachments, extract their contents, assess legal validity or approve a commercial document. File existence and readability should come from ordinary application checks.
This is an original synthetic demonstration adapted from the cited workflow discussion. It does not reproduce the source application or verify its performance, adoption or security claims.

Use the sample to understand the rule, then test your own input for document intake checklist.

Provide the requirements and submission notes.

Open the decision rules and keep the intended distinction between Required evidence present, Required part missing, Reference unresolved, Checklist unclear, 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 Required part missing, Checklist unclear, 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 parts present | Required evidence present | 98.0% |
| Reviewer note absent | Required part missing | 99.0% |
| Attachment mentioned only | Reference unresolved | 99.0% |
| Undefined requirement | Checklist unclear | 89.0% |
| Optional part not required | Required evidence present | 99.0% |
| No checklist | Required part missing | 71.0% |
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. 5 of 6 results matched the authored fixture expectations. For No checklist, the expectation was Need evidence; the observed choice was Required part 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 checks presence under the supplied checklist. Correctness, authenticity and substantive review are separate.
No. Ordinary code should check file presence and access. The model interprets those verified notes against the semantic checklist.
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.