Canvas shape references: 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 transcript, shape descriptions and recorded pointing or selection context. Resolve a supplied object ID, keeping missing or conflicting references unresolved.
A canvas editor needs an object ID before it can move or restyle anything. Natural references such as this one depend on state outside the sentence, including what the person pointed at while speaking.
The examples supply that state as text records. A real interface would capture pointer events, selection and timestamps separately, then validate the returned ID against its current canvas.
The linked Jev Canvas project uses choices over shapes and contextual references. This page demonstrates target resolution only. It does not listen to speech, track hands, inspect an image or edit a canvas.
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 canvas shape references.

Provide the transcript, shape descriptions and recorded pointing or selection context.

Open the decision rules and keep the intended distinction between A, B, C, Unresolved. 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 uncertain or unexpected results before using a result. The label is a decision about the supplied information, not an executed action.
| Original input | Observed label | Leading probability |
|---|---|---|
| Named appearance | B | 95.0% |
| Pointing at speech time | C | 98.0% |
| Repeat reference | A | 99.0% |
| Selection reference | B | 100% |
| No pointing record | Unresolved | 68.0% |
| Two matching shapes | Unresolved | 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
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. You edit a transcript and explicit state records. Any speech transcription or hand tracking would be a separate part of an application.
The demonstration rule gives this one a specific pointing reference. A selected object is not evidence that the person pointed at it when speaking.
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.