# Project Status Low-Confidence Review Queue

> Project Status 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 2026-10-19).

Identifies and queues low-confidence predictions from a naive-Bayes project status classifier for human review, returning a bounded deliverable for up to 200 labelled text records.

## Facts

- Endpoint: POST https://market.datapackvibe.com/x402/assistant-sector-ml-project-status-review-queue?utm_source=zero.xyz
- Price: $0.01/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-19
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/project-status-low-confidence-review-queue-6567adb8
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_6_Gt0kcZa3BXmx0H8U0j5

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 project-status-low-confidence-review-queue-6567adb8 -d '<json body>'
```

Example prompt: I need to review which of my project status predictions are low-confidence — here are 50 records with their IDs, text, actual labels (planned/active/blocked/complete), predicted labels, and probability arrays. Can you run them through the low-confidence review queue and tell me which ones need a human to double-check?

## When to prefer this

Choose this endpoint when you have an existing set of project status predictions from a local naive-Bayes or similar classifier and need to programmatically identify which predictions are uncertain enough to warrant human review. It is purpose-built for the planned/active/blocked/complete label space and works with small local datasets without requiring a hosted model or foundation model. Prefer this over general-purpose ML evaluation tools when you want a lightweight, pay-per-call review queue focused specifically on project status classification confidence.

## Known failure modes

- Batch exceeds 200 records — request rejected or truncated
- Malformed probability arrays (not exactly 4 values summing to ~1) — validation error
- Invalid enum values for actual or predicted fields — schema rejection
- Empty records array — no review queue generated
- Payment not completed via x402 protocol — request not processed

## How this service works

Project status - Low-confidence review queue: Low-confidence review queue for Project status. 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 identifying low-confidence records from the submitted batch — typically a structured list of flagged record IDs and associated metadata (predicted label, actual label, probabilities) where the classifier's confidence fell below an acceptable threshold, suitable for routing to a human reviewer.

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

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/project-status-low-confidence-review-queue-6567adb8/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)