Popular repositories Loading

1. AI-Ethics-Auditor AI-Ethics-Auditor Public

   Production-grade platform for auditing AI/ML models for fairness, explainability, and robustness. JWT auth, SHA-256 model verification, paginated REST API, dark web dashboard. 30/30 tests passing.

   Python

2. adversarial-immune-system adversarial-immune-system Public

   Self-learning system that blocks adversarial prompts using activation-space geometry. 100% detection rate, F1=0.926, ROC-AUC=0.977. No pattern matching - detects structural fragility.

   Python

3. prompt-lint prompt-lint Public

   A linter for LLM system prompts - detects contradictions, injection risks, security antipatterns, and more.

   Python

4. fairness-drift fairness-drift Public

   Continuous Temporal Fairness Orchestrator using an O(1) Detection Triangle (F-ADWIN, F-EWMA, Page-Hinkley). Designed to statistically prove and detect sudden shifts, gradual data poisoning, and mon…

   Python