Jev midrun message kind

Classify a message that arrives while an agent is working

Compare the incoming message with the current goal and action. Distinguish a correction, added work and a side question; the result describes intent rather than interrupting or executing work.

Is the user correcting, adding or asking?Live API
Try an example

The user identifies a wrong ongoing action.

Source: API-collected workflow inspiration; original demonstration below

Source and example scope

The API-collected roundup links the cited project. Its pinned source defines the decision adapted here. These synthetic inputs and explicit missing-context outcomes are original to this demonstration. The window returns a label; it does not run the linked application, execute its downstream actions or verify its performance claims.

Compare userMessage with the currentAction and goal.

Decision rules and labels
CorrectionAdded workSide questionOtherNeed context
A real response from the Jev API

Edit any input to explore a different decision. No request is sent until you run it.

Your next decision

Not run yet

Choose an example or write your own input, then run it to see the ranked choices.

Results are not stored by this site. Your submitted text is processed by the API provider.

The decision

A correction to the current action differs from work requested for later.

The staging preset includes a concrete mismatch: the agent is about to edit production even though staging is the goal. The report preset instead leaves the test run intact and asks for a later deliverable.

A message can contain both. The mixed preset first tells the agent to stop the wrong edit, so Correction takes priority over the requested summary. That priority is part of this original demonstration’s displayed rule.

The source project maps message classes into its own routing policy. This window stops at classification: it neither interrupts another agent nor queues commands. Missing current-action context remains unresolved.

The API-collected roundup links the cited project. Its pinned source defines the decision adapted here. These synthetic inputs and explicit missing-context outcomes are original to this demonstration. The window returns a label; it does not run the linked application, execute its downstream actions or verify its performance claims.

Illustration of choosing one tool from the supplied alternatives
Illustration of the decision flow.
How it works

Try the example in four steps

Use the sample to understand the rule, then test your own input for messages during agent work.

  1. Illustration of adding the text or evidence for a decision

    Add the information to check

    Compare the incoming message with the current goal and action.

  2. Illustration of defining a question and its allowed choices

    Confirm the question and labels

    Open the decision rules and keep the intended distinction between Correction, Added work, Side question, Other, Need context. Edit the choices only when testing your own rule.

  3. Illustration of sending one input and comparing the returned choices

    Run the decision

    Use a sample first, then try your own input. Compare the selected label with the other returned probabilities.

  4. Illustration of reviewing a result and an uncertain item before using it

    Review the next step

    Review Need context before using a result. The label is a decision about the supplied information, not an executed action.

Our API test

Recorded API observations

Recorded API observations
Original inputObserved labelLeading probability
Correct the targetCorrection100%
Add a later reportAdded work100%
Unrelated questionSide question100%
Simple acknowledgmentOther100%
Correct and addCorrection99.0%
Missing actionNeed context52.0%

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

Understand the results

Replay the saved synthetic requests to inspect each recorded decision

Replay the saved synthetic requests to inspect each recorded decision. Authored expectations and model disagreements remain available; these examples are not a production accuracy benchmark.

Using the result in your application

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.

Runnable offline example

Run one sample, then inspect the local packet

View the code example
Single-request replay
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.json
Scope of this demo

Where this decision fits

Useful scope

  • Testing message routing rules with synthetic exchanges
  • Separating corrections from later additions
  • Making mixed-message priority explicit

Outside this demonstration

  • Interrupting another agent from this page
  • Inferring the active task without context
  • Treating the label as an execution permission
Example FAQ

Questions about midrun message kind.

Input scope and downstream handling determine what this judgment means.

Does Correction stop a running action?

No. A controller must separately decide whether and how to stop it. This page returns only the text classification.

Where does a question about the current task go?

Under this narrow catalog it is Other unless it corrects the action or adds work. Side question is reserved for an unrelated question.

Is the leading probability an accuracy measurement?

No. It is one value in the returned distribution over this request’s choices. The six synthetic fixtures have authored expectations; they do not measure production accuracy.

What happens on an API failure?

The window displays an error without a fresh judgment. An incomplete-input outcome describes the supplied evidence and is not used as a fallback for provider failure.

Try your own midrun message kind input

Edit the input above, or open the Playground to explore another decision.