← ClaudeAtlas

anomaly-scanlisted

Scan all connected marketing platforms for statistically significant deviations from stored baselines — traffic drops, CPA spikes, deliverability collapse, budget overruns, or unexpected wins — classified critical/warning/info with probable causes, correlation to recent changes, and recommended actions. Triggers on "/digital-marketing-pro:anomaly-scan", "why did our CPA spike", "did anything weird happen this week", "check for anomalies", "our conversions suddenly dropped". Runs performance-monitor.py for baselines and detection, correlates flags against execution-tracker.py history and the diagnostic framework in skills/analytics-insights/anomaly-diagnosis.md, and persists critical findings as insights via campaign-tracker.py. Reads the brand profile.
teachskillofskills-ai/DigitalMarketingPro-techshu · ★ 0 · Data & Documents · score 70
Install: claude install-skill teachskillofskills-ai/DigitalMarketingPro-techshu
# /digital-marketing-pro:anomaly-scan ## Purpose Scan all connected marketing platforms for anomalies — statistically significant deviations from established baselines that could indicate problems (traffic drops, CPA spikes, deliverability collapse, budget overruns) or opportunities (viral content, conversion rate improvements, unexpected channel growth). Designed to catch issues early, before they compound into costly problems, and to surface wins worth amplifying. ## Input Required The user must provide (or will be prompted for): - **Sensitivity level**: Strict (flags deviations >1.5 standard deviations from baseline), normal (>2 std dev), or relaxed (>3 std dev). Defaults to normal - **Time period**: The window to scan for anomalies — today, last 3 days, last 7 days, last 30 days, or custom range. Defaults to last 7 days - **Platforms** (optional): Specific platforms to focus the scan on (e.g., "Google Ads and Meta only"). If omitted, all connected platforms are scanned - **Metrics focus** (optional): Specific metrics to prioritize (e.g., "CPA and conversion rate only"). If omitted, all available metrics are evaluated - **Baseline period** (optional): Custom baseline for comparison instead of the default. Defaults to the rolling 30-day average maintained by performance-monitor.py - **Exclude known events** (optional): List of known events to filter out (e.g., "Black Friday sale", "site migration on Jan 15") so expected deviations are not flagged as anomali