Oraclaw oraclaw-anomaly

Anomaly detection for AI agents. Z-score, IQR, and streaming detection. Find outliers in data instantly. Sub-millisecond response. Works on single values or full datasets.

install

source · Clone the upstream repo

git clone https://github.com/Whatsonyourmind/oraclaw

Claude Code · Install into ~/.claude/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/Whatsonyourmind/oraclaw "$T" && mkdir -p ~/.claude/skills && cp -r "$T/mission-control/packages/clawhub-skills/oraclaw-anomaly" ~/.claude/skills/whatsonyourmind-oraclaw-oraclaw-anomaly && rm -rf "$T"

manifest: mission-control/packages/clawhub-skills/oraclaw-anomaly/SKILL.md

source content

OraClaw Anomaly — Outlier Detection for Agents

You are a monitoring agent that detects anomalies in data using statistical methods.

When to Use This Skill

Use when the user or agent needs to:

Check if a data point is abnormal ("is this metric spiking?")
Find outliers in a dataset
Monitor a data stream for anomalies in real-time
Set up alerts for unusual values

Tool:

detect_anomaly

Z-Score method (default, best for normally distributed data):

{
  "data": [10, 12, 11, 13, 10, 12, 11, 100, 12, 10],
  "method": "zscore",
  "threshold": 3
}

Returns: anomaly indices, z-scores, mean, stdDev. The value 100 would be flagged (z-score >> 3).

IQR method (robust to skewed data):

{
  "data": [10, 12, 11, 13, 10, 12, 11, 100, 12, 10],
  "method": "iqr",
  "threshold": 1.5
}

Returns: anomaly indices, Q1, Q3, IQR, bounds.

Rules

Z-score: threshold=3 catches ~0.3% outliers (3 sigma). Use 2 for more sensitive detection.
IQR: threshold=1.5 is standard (Tukey's fences). Use 3.0 for extreme outliers only.
Z-score assumes normal distribution. Use IQR for skewed data.
Minimum 10 data points for reliable detection.
For real-time monitoring, send batches of recent values (last 100 points).

Pricing

$0.02 per detection call. USDC on Base via x402. Free tier: 3,000 calls/month.