# Aient Anomaly Detection

> Aient Anomaly Detection is a paid API for AI agents from aient.ai, paid per call via x402, $0.010000/call, status unknown (last checked 2026-09-15).

Detects statistical anomalies in error rate, latency, or throughput for your services using windowed regression — the same algorithm Aient uses internally to identify problems.

## Facts

- Endpoint: POST https://aient.ai/x402/anomalies
- Price: $0.010000/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/aient-ai-09b34ee7
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap__GTnQwjJgc_Yn_S-Aw0R4

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 aient-ai-09b34ee7 -d '<json body>'
```

Example prompt: Can you run anomaly detection on my Aient-monitored services and flag any statistical outliers in error rate or latency from the last 2 hours, grouped by service?

## When to prefer this

Use this endpoint when you need statistically rigorous anomaly detection rather than simple threshold alerts — specifically when you want the same algorithm Aient uses internally to spawn problems. Prefer this over raw trace queries when you want pre-computed regression-based outlier detection across error rate, latency, or throughput dimensions grouped by service or operation.

## Known failure modes

- Invalid or missing ingest/API key — 401 Unauthorized
- No telemetry data ingested for the requested time range — empty results or 404
- Invalid metric type or grouping field — 400 Bad Request
- Insufficient data points for regression — model may return no anomalies or a specific error
- Rate limiting or quota exceeded — 429 Too Many Requests
- Malformed request body — 422 Unprocessable Entity

## How this service works

Detect statistical anomalies across error rate, latency, or throughput grouped by service or operation — the same algorithm Aient uses internally to spawn problems. Pricier than raw queries because it runs windowed regression.

## Output

Returns a list of detected anomalies grouped by service or operation, each with the metric type (error_rate, latency, or throughput), the anomalous window, an anomaly score, and contextual statistics from the windowed regression model.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "since": {
   "type": "string",
   "format": "date-time"
  },
  "until": {
   "type": "string",
   "format": "date-time"
  },
  "metric": {
   "enum": [
    "error_rate",
    "duration_p99",
    "throughput"
   ],
   "type": "string"
  },
  "groupBy": {
   "enum": [
    "service",
    "operation"
   ],
   "type": "string"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/aient-ai-09b34ee7/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from aient.ai](https://www.zero.xyz/host/aient.ai/llms.txt)