# Support Intent Low-Confidence Review Queue

> Support Intent Low-Confidence Review Queue is a paid API for AI agents from market.datapackvibe.com, paid per call via x402, $0.01/call, status unknown (last checked 2026-10-12).

Identifies and surfaces low-confidence support ticket classifications (billing, technical, account, other) from a naive-Bayes model for human review.

## Facts

- Endpoint: POST https://market.datapackvibe.com/x402/assistant-sector-ml-support-intent-review-queue?utm_source=zero.xyz
- Price: $0.01/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-12
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/support-intent-low-confidence-review-queue-f8746fd3
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_lVKIJ6ZxSJ_bxHHzQmLPy

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 support-intent-low-confidence-review-queue-f8746fd3 -d '<json body>'
```

Example prompt: I have up to 200 labeled support ticket records — each with an ID, the text, the actual intent label (billing, technical, account, or other), the predicted label, and four class probabilities — can you run them through the low-confidence review queue so I can see which ones the classifier was uncertain about and need human review?

## When to prefer this

Choose this endpoint when you have a labeled support-intent dataset (up to 200 records) and want to identify which naive-Bayes predictions are low-confidence for human review, without needing a hosted model or foundation-model training. Ideal for QA loops before classifier retraining or auditing routing accuracy on small local datasets.

## Known failure modes

- More than 200 records submitted — endpoint enforces a hard cap
- Probabilities array does not have exactly 4 elements per record — schema validation failure
- Label values outside the allowed enum (billing, technical, account, other) cause rejection
- Record ID missing or empty string triggers validation error
- Text field exceeds 1000-character limit per record

## How this service works

Support intent - Low-confidence review queue: Low-confidence review queue for Support intent. Up to 200 supplied labelled text records. Local dataset tools, evaluation and small naive-Bayes baselines; no hosted model or foundation-model training. One of 25 profiles or calculations across 20 shared operations in this sector. 0.01 USDC includes one bounded low-confidence review queue deliverable.

## Output

A bounded deliverable (up to 200 records processed) identifying which submitted support text records fell below confidence thresholds in the naive-Bayes classifier, surfaced as a low-confidence review queue with flagged record IDs, their predicted and actual labels, and associated probability scores — ready for human review or relabeling.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "records": {
   "type": "array",
   "items": {
    "type": "object",
    "required": [
     "id",
     "text",
     "actual",
     "predicted",
     "probabilities"
    ],
    "properties": {
     "id": {
      "type": "string",
      "default": "",
      "maxLength": 80,
      "minLength": 1
     },
     "text": {
      "type": "string",
      "default": "",
      "maxLength": 1000
     },
     "actual": {
      "enum": [
       "billing",
       "technical",
       "account",
       "other"
      ],
      "type": "string"
     },
     "predicted": {
      "enum": [
       "billing",
       "technical",
       "account",
       "other"
      ],
      "type": "string"
     },
     "probabilities": {
      "type": "array",
      "items": {
       "type": "number",
       "default": 0,
       "maximum": 1,
       "minimum": 0
      },
      "default": [],
      "maxItems": 4,
      "minItems": 4
     }
    },
    "additionalProperties": false
   },
   "default": [
    {
     "id": "example-0-0",
     "text": "billing sample document number 0 for this supplied dataset.",
     "actual": "billing",
     "predicted": "billing",
     "probabilities": [
      0.8,
      0.06666666666666667,
      0.06666666666666667,
      0.06666666666666667
     ]
    },
    {
     "id": "example-0-1",
     "text": "billing sample document number 1 for this supplied dataset.",
     "actual": "billing",
     "predicted": "billing",
     "probabilities": [
      0.8,
      0.06666666666666667,
      0.06666666666666667,
      0.06666666666666667
     ]
    },
    {
     "id": "example-0-2",
     "text": "billing sample document number 2 for this supplied dataset.",
     "actual": "billing",
     "predicted": "billing",
     "probabilities": [
      0.8,
      0.06666666666666667,
      0.06666666666666667,
      0.06666666666666667
     ]
    },
    {
     "id": "example-1-0",
     "text": "technical sample document number 0 for this supplied dataset.",
     "actual": "technical",
     "predicted": "technical",
     "probabilities": [
      0.06666666666666667,
      0.8,
      0.06666666666666667,
      0.06666666666666667
     ]
    },
    {
     "id": "example-1-1",
     "text": "technical sample document number 1 for this supplied dataset.",
     "actual": "technical",
     "predicted": "technical",
     "probabilities": [
      0.06666666666666667,
      0.8,
      0.06666666666666667,
      0.06666666666666667
     ]
    },
    {
     
… (truncated)
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/support-intent-low-confidence-review-queue-f8746fd3/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from market.datapackvibe.com](https://www.zero.xyz/host/market.datapackvibe.com/llms.txt)