# Task Type Low-Confidence Review Queue

> Task Type 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 predictions from a naive-Bayes task-type classifier for human review, given up to 200 labelled text records with actual/predicted labels and class probabilities.

## Facts

- Endpoint: POST https://market.datapackvibe.com/x402/assistant-sector-ml-task-type-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/task-type-low-confidence-review-queue-0118a664
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_9Rb3WFu7XYAaOf2YGsBy-

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 task-type-low-confidence-review-queue-0118a664 -d '<json body>'
```

Example prompt: I have 200 labelled text records — each with an ID, the text, the actual task type (research, writing, analysis, or implementation), the predicted task type, and the four class probabilities — can you run them through the low-confidence review queue to surface which ones I should manually check?

## When to prefer this

Choose this endpoint when you have a local text dataset labelled with task types (research, writing, analysis, implementation) and need to identify which predictions from a naive-Bayes model are uncertain enough to warrant human review. It is ideal for data quality audits, pre-retraining cleaning, and lightweight evaluation workflows where no hosted model or foundation-model training is needed.

## Known failure modes

- Fewer than 1 or more than 200 records supplied — request rejected
- Records missing required fields (id, text, actual, predicted, probabilities) — validation error
- Probabilities array does not contain exactly 4 values summing approximately to 1 — schema rejection
- Invalid enum values for actual or predicted fields — rejected
- Empty records array returns no meaningful review queue

## How this service works

Task type - Low-confidence review queue: Low-confidence review queue for Task type. 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 review queue deliverable listing up to 200 records identified as low-confidence by the naive-Bayes classifier, with associated metadata (IDs, text, actual vs predicted labels, probabilities) to guide human review and relabelling.

## 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": [
       "research",
       "writing",
       "analysis",
       "implementation"
      ],
      "type": "string"
     },
     "predicted": {
      "enum": [
       "research",
       "writing",
       "analysis",
       "implementation"
      ],
      "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": "research sample document number 0 for this supplied dataset.",
     "actual": "research",
     "predicted": "research",
     "probabilities": [
      0.8,
      0.06666666666666667,
      0.06666666666666667,
      0.06666666666666667
     ]
    },
    {
     "id": "example-0-1",
     "text": "research sample document number 1 for this supplied dataset.",
     "actual": "research",
     "predicted": "research",
     "probabilities": [
      0.8,
      0.06666666666666667,
      0.06666666666666667,
      0.06666666666666667
     ]
    },
    {
     "id": "example-0-2",
     "text": "research sample document number 2 for this supplied dataset.",
     "actual": "research",
     "predicted": "research",
     "probabilities": [
      0.8,
      0.06666666666666667,
      0.06666666666666667,
      0.06666666666666667
     ]
    },
    {
     "id": "example-1-0",
     "text": "writing sample document number 0 for this supplied dataset.",
     "actual": "writing",
     "predicted": "writing",
     "probabilities": [
      0.06666666666666667,
      0.8,
      0.06666666666666667,
      0.06666666666666667
     ]
    },
    {
     "id": "example-1-1",
     "text": "writing sample document number 1 for this supplied dataset.",
     "actual": "writing",
     "predicted": "writing",
     "probabilities": [
      0.06666666666666667,
      0.8,
      0.06666666666666667,
      0.06666666666666667
     ]
   
… (truncated)
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/task-type-low-confidence-review-queue-0118a664/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)