Evally
Start free
Product

User intent classifier

What does the user want, and are they telling the app to do something right now?

Classifies what a user typed or said into your intents, and separately checks whether it's a command for the software rather than chatter or dictation, which is how always-listening assistants avoid acting on everything they hear. Unclear requests get a clarifying question. Replace the intents with your own.

Use this judgeCustomize it

Try it live

Real calls to Jev. Pick an example or paste your own.

Examples
60
No verdict yet

Pick an example or paste your own input, then run it.

Under the hood

Every question and rule, in the open. Fork it to change any of them.

What it asks

Reads utterance, then asks 2 questions in one call.

  1. Q1IntentPick one

    What does the user want?

    Ask a questionReport a problemRequest a featureBuy or upgradeCancelGive feedbackGive a commandSmall talkUnclear
  2. Q2Instruction to the appYes / no

    Is the utterance an instruction for the software to do something now, as opposed to talking, thinking aloud, or dictating text?

How it decides

Rules run in code, top to bottom. The first one that matches gives the verdict.

  1. 1
    If Intent: confidence < 0.6Ask to clarify
  2. 2
    If Intent is UnclearAsk to clarify
  3. 3
    Otherwisethe answer to “Intent”
  4. ·
    DefaultAsk to clarify

Call it from your code

Every template runs as-is with an API key, no setup: POST /api/v1/judges/intent-classifier/run.

curl -X POST https://tryevally.com/api/v1/judges/intent-classifier/run \  -H "Authorization: Bearer $EVALLY_API_KEY" \  -H "Content-Type: application/json" \  -d '{  "input": {    "utterance": "Turn off the living room lights and set the thermostat to 20"  }}'

Seen in the wild

Community builds on Jev that use this pattern. Not affiliated with Evally.

More in Product