Chat reading intent: six recorded examples
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
Provide a chat message and relevant thread context. Distinguish a practical answer, information request and suggestion, keeping mixed purposes for review.
Reading intent is a preference for which messages to surface. A host looking for questions needs a different selection from a host looking for answers to a technical problem.
This taxonomy keeps practical help separate from feedback about the stream. The same sentence may serve a different role in another thread, so supply the question it answers when that context matters.
Jev Chat for Twitch filters messages by selected reading categories. This original text example uses a smaller taxonomy and does not connect to Twitch, delete messages or judge whether someone should be banned.
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 chat reading intent.

Provide a chat message and relevant thread context.

Open the decision rules and keep the intended distinction between Help, Question, Feedback, Other, 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 |
|---|---|---|
| A direct viewer question | Question | 100% |
| Answering a shortcut question | Help | 100% |
| Readability suggestion | Feedback | 99.0% |
| A greeting | Other | 100% |
| Two equal purposes | Review | 91.0% |
| Unresolved reply | Review | 92.0% |
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.
No. It only means the message has another purpose, such as greeting the host. The labels do not measure politeness or safety.
Yes. This single-choice demonstration sends equally central competing intents to Review. A different product could ask separate questions for each intent.
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.