PDF text-layer steganography toolkit — hide messages in PDFs that look identical to humans but are extractable by software. For watermarking, AI ingestion-pipeline testing, CTFs, and research.

Empirical dual-use risk assessment of protein language models (ESM-2) and structure-based design tools (ProteinMPNN)

Dual-Use Ambiguity Casebook: 30 structured cases at the AI-era biology dual-use decision boundary

Biological dual-use content classifier using Constitutional Classifiers methodology — biosafety constitution, synthetic data pipeline, DeBERTa-v3-base classifier

Pharmakon-Eval: dual-use LLM evaluation via Pharmakon Separability Index (PSI) and Dose-Response Curve (DRC)

30 machine-readable constitutional rules for biological dual-use content across 6 bio domains. JSON format for Constitutional Classifier pipeline integration.