# Intent Detection

> Intent Detection is a paid API for AI agents from x402.agentindex.world, paid per call via x402, $0.002/call, status unknown (last checked 2026-10-29).

Classifies text as a question, request, complaint, compliment, or other, returning a confidence score and alternate label within 4.5 seconds

## Facts

- Endpoint: GET https://x402.agentindex.world/intent?utm_source=zero.xyz
- Price: $0.002/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-29
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/intent-detection-b06745c1
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_T2mCuMR6BgQWQkPYoS69o

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability intent-detection-b06745c1
```

Example prompt: Classify this message by intent and give me a confidence score: 'I've been waiting three weeks and my order still hasn't arrived — this is unacceptable.'

## When to prefer this

Choose this endpoint when you need fast, structured intent classification with a confidence score and alternate label, particularly for customer-facing text like support tickets, chat messages, or user reviews. It is ideal when you need guaranteed sub-4.5s latency thanks to its dual-provider architecture (Jev + Groq fallback), and when you want a pay-per-call model with no subscription overhead. Prefer it over general-purpose LLM prompting when you want a consistent, schema-bound output for downstream routing logic.

## Known failure modes

- Text exceeding 4000 characters may be rejected or truncated
- Both Jev and Groq fallback unavailable — call may fail after timeout
- Ambiguous or very short text may produce low-confidence or inaccurate classifications
- Non-English text may yield unreliable results depending on model support
- Network timeout beyond the 4.5s guaranteed SLA if both providers are degraded

## How this service works

Intent detection: classify text as a question, request, complaint, compliment, or other, with a confidence score and an alternate label. Powered by Jev, with an automatic Groq-hosted LLM fallback if Jev is slow or unavailable - under 4.5s guaranteed. Try GET /intent/sample.

## Output

Returns a primary intent label (question, request, complaint, compliment, or other), a confidence score between 0 and 1, and an alternate label suggesting the next most likely intent category.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "text": {
   "type": "string",
   "description": "Text to classify, up to 4000 characters."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "input": "Could someone take a look at why my last two orders were both delayed?",
  "label": "request",
  "engine": "jev",
  "probability": 0.55,
  "alternate_label": "question",
  "alternate_probability": 0.29
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/intent-detection-b06745c1/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from x402.agentindex.world](https://www.zero.xyz/host/x402.agentindex.world/llms.txt)