Recipe catalog / trip-purpose-kind

Identify the purpose of a trip

What purpose does the traveler's message express for the trip: business, leisure, a family visit, medical care, relocation, or attending an event?

A travel assistant needs to route a traveler's message to the right search defaults, policy, or recommendations based on why they are traveling, using only what the message and any prior context say.

Explore this recipe interactively · Source and implementation guide

Use trip-purpose-kind in TypeScript

Install with npm install jev-recipes. Requires Node.js 22.9 or newer and ES modules. Set TYPESAFE_API_KEY in your server environment for live calls, which send input to TypeSafe and use API quota. See the installation guide.

import { tripPurposeKind } from 'jev-recipes/trip-purpose-kind';

const result = await tripPurposeKind({
  "message": "We are moving to Lisbon for good at the end of October. My partner's job transfers her there permanently and the kids start school in January. I need one-way flights for the four of us and somewhere to stay for the first three weeks until our long-term lease begins on 20 November.",
  "context": "Traveler previously asked this assistant about shipping household goods overseas.",
  "minConfidence": 0.8
});
console.log(result);

Input contract

{
  "input": {
    "$schema": "https://json-schema.org/draft/2020-12/schema",
    "type": "object",
    "properties": {
      "message": {
        "type": "string"
      },
      "context": {
        "type": "string"
      },
      "minConfidence": {
        "type": "number",
        "minimum": 0,
        "maximum": 1
      }
    },
    "required": [
      "message"
    ]
  },
  "result": {
    "$schema": "https://json-schema.org/draft/2020-12/schema",
    "type": "object",
    "properties": {
      "model": {
        "type": "string"
      },
      "usage": {
        "type": "object",
        "properties": {
          "input_tokens": {
            "type": "integer",
            "minimum": 0,
            "maximum": 9007199254740991
          },
          "output_tokens": {
            "type": "integer",
            "minimum": 0,
            "maximum": 9007199254740991
          }
        },
        "required": [
          "input_tokens",
          "output_tokens"
        ],
        "additionalProperties": false
      },
      "status": {
        "type": "string",
        "enum": [
          "ready",
          "review"
        ]
      },
      "verdict": {
        "type": "string",
        "enum": [
          "business",
          "leisure",
          "family_visit",
          "medical",
          "relocation",
          "event",
          "unclear"
        ]
      },
      "confidence": {
        "type": "number",
        "minimum": 0,
        "maximum": 1
      },
      "probabilities": {
        "type": "object",
        "propertyNames": {
          "type": "string",
          "enum": [
            "business",
            "leisure",
            "family_visit",
            "medical",
            "relocation",
            "event",
            "unclear"
          ]
        },
        "additionalProperties": {
          "type": "number",
          "minimum": 0,
          "maximum": 1
        },
        "required": [
          "business",
          "leisure",
          "family_visit",
          "medical",
          "relocation",
          "event",
          "unclear"
        ]
      }
    },
    "required": [
      "model",
      "usage",
      "status",
      "verdict",
      "confidence",
      "probabilities"
    ],
    "additionalProperties": false
  }
}

Saved example result

This hand-authored response demonstrates the contract. It is not a model accuracy measurement. Run it without an API key: npx jev-recipes demo trip-purpose-kind.

{
  "model": "demo-fixture",
  "usage": {
    "input_tokens": 0,
    "output_tokens": 0
  },
  "status": "ready",
  "verdict": "relocation",
  "confidence": 0.9,
  "probabilities": {
    "business": 0.02,
    "leisure": 0.01,
    "family_visit": 0.02,
    "medical": 0.01,
    "relocation": 0.9,
    "event": 0.01,
    "unclear": 0.03
  }
}

Evaluation evidence

Use the evaluation guide to measure this decision on your own labeled cases.

Limitations

• Names the primary purpose the wording expresses. Mixed trips return the dominant purpose or unclear; it does not split a trip into segments.
• Does not verify the stated purpose or infer one the traveler did not express. Corporate travel policy and eligibility decisions belong in application code.

Related recipes

• turn-intent: Use turn-intent to classify what a message is doing conversationally, such as requesting or correcting, rather than why the person is traveling.
• buying-intent: Use buying-intent to judge how close the traveler is to booking; this recipe identifies only the reason for the trip.