anomaly-detection
Anomaly Detection
Rule-based anomaly detection with cooldowns and error pattern tracking.
When to Use This Skill
- Detecting slow job degradation before failures
- Tracking error rate creep over time
- Identifying repeated error patterns
- Preventing alert fatigue with cooldowns
Core Concepts
Production systems fail in subtle ways - jobs getting slower, error rates creeping up, same errors repeating. The solution:
- Configurable rules with severity levels
- Cooldown periods to prevent alert storms
- Error pattern tracking for repeated failures
- Violation decay to reward recovery
More from dadbodgeoff/drift
sse-streaming
Implement Server-Sent Events (SSE) for real-time updates with automatic reconnection and heartbeats. Use when building live dashboards, notifications, progress indicators, or any feature needing server-to-client push.
78oauth-social-login
Implement OAuth 2.0 social login with Google, GitHub, and other providers. Handles token exchange, user creation, and account linking.
48multi-tenancy
Implement multi-tenant architecture with tenant isolation, data separation, and per-tenant configuration. Supports shared database and schema-per-tenant models.
45deduplication
Event deduplication with canonical selection, reputation scoring, and hash-based grouping for multi-source data aggregation. Handles both ID-based and content-based deduplication.
43fuzzy-matching
Multi-stage fuzzy matching pipeline for entity reconciliation. PostgreSQL trigram pre-filter, salient overlap check, and multi-factor similarity scoring.
40webhook-security
Implement secure webhook handling with signature verification, replay protection, and idempotency. Use when receiving webhooks from third-party services like Stripe, GitHub, Twilio, or building your own webhook system.
37