Attachment relevance: six recorded examples
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
Provide the target record, message and attachment preview. Compare an attachment worth reading with unrelated material or a preview too sparse to decide.
An attachment filename alone is a weak signal. This decision compares the supplied message and preview with one target record, so a valid invoice for another customer can still be skipped.
Read means the supplied description contains relevant information. Review preserves uncertainty when a file might matter but its contents are unknown. Neither outcome checks malware, reads a PDF or grants access.
This is an independently authored text demonstration inspired by the linked X discussion. Its rules and synthetic inputs are ours. It does not reproduce the author’s application, verify reported outcomes, or perform an external action.

Use the sample to understand the rule, then test your own input for attachment relevance.

Provide the target record, message and attachment preview.

Open the decision rules and keep the intended distinction between Read, Skip, Review. 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 Review before using a result. The label is a decision about the supplied information, not an executed action.
| Original input | Observed label | Leading probability |
|---|---|---|
| Quoted work | Read | 100% |
| Signature image | Skip | 95.0% |
| Unreadable scan | Review | 93.0% |
| Different customer | Skip | 99.0% |
| Payment receipt | Read | 98.0% |
| Misleading filename | Skip | 86.0% |
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. 6 of 6 results matched the authored fixture expectations.Verified
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
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.
# Offline: inspect the downloaded saved observations. No API key is needed.
python preview_choice_workflow.py samples.json saved-api-results.json
# Optional live request: install the SDK and set TYPESAFE_API_KEY privately.
npm install @typesafe-ai/sdk@0.6.0
node run-choice-sample.mjs samples.json case-1
# The live response prints to the terminal; it does not update saved-api-results.json.The input contract and these limits define what the result means.
Yes, but a vague filename may lead to Review. Include the email context and target record; the model cannot inspect bytes that you have not supplied.
No. It means relevant to the target record. File scanning, access checks and a sandbox belong in the application that opens the attachment.
It describes the distribution over the supplied choices for this request. These six synthetic examples are functional tests with authored expectations, not a production accuracy benchmark.
No. The demo displays a label and distribution. The downloadable consumer produces a pending preview packet. Application code and a separate authorization policy would control any real action.
Edit the example above to see how a changed fact affects the decision, or open the Playground for another task.