Recipe catalog / escalation-wording

Check whether a message asks for escalation

Does message explicitly ask for escalation to a higher tier, a manager, or on-call engineering?

You are routing inbound support messages or internal chat and need to catch explicit escalation requests so they reach a supervisor, a higher support tier, or the on-call engineer.

Explore this recipe interactively · Source and implementation guide

Use escalation-wording 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 { escalationWording } from 'jev-recipes/escalation-wording';

const result = await escalationWording({
  "message": "I have been going back and forth with support for a week and the billing error is still not fixed. Please escalate this to your engineering team or have a supervisor contact me directly today.",
  "minConfidence": 0.8
});
console.log(result);

Input contract

{
  "input": {
    "$schema": "https://json-schema.org/draft/2020-12/schema",
    "type": "object",
    "properties": {
      "message": {
        "type": "string"
      },
      "minConfidence": {
        "type": "number",
        "minimum": 0,
        "maximum": 1
      }
    },
    "required": [
      "message"
    ]
  },
  "result": {
    "$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"
        ]
      },
      "probability": {
        "type": "number",
        "minimum": 0,
        "maximum": 1
      },
      "confidence": {
        "type": "number",
        "minimum": 0,
        "maximum": 1
      },
      "verdict": {
        "type": "string",
        "enum": [
          "requested",
          "absent"
        ]
      }
    },
    "required": [
      "model",
      "usage",
      "status",
      "probability",
      "confidence",
      "verdict"
    ],
    "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 escalation-wording.

{
  "model": "demo-fixture",
  "usage": {
    "input_tokens": 0,
    "output_tokens": 0
  },
  "status": "ready",
  "probability": 0.96,
  "confidence": 0.96,
  "verdict": "requested"
}

Evaluation evidence

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

Limitations

• Detects an explicit request, not a need. A frustrated message that deserves escalation but does not ask for it yields absent.
• Does not decide whether the request should be granted or who the right recipient is. Routing rules belong in application code.

Related recipes

• urgency-signal: Use urgency-signal to grade how urgent a message sounds, whether or not it asks for escalation.
• handoff: Use handoff to decide whether a conversation should move to a human at all, rather than whether the writer asked for a higher tier.