Cryptographically verifiable audit infrastructure for AI agents. AgentOps Replay produces tamper-evident, hash-chained event logs that prove what an agent did, in what order, with no possibility of post-hoc modification. Built for multi-agent safety research, forensic audit, and regulatory accountability where observability alone is not enough.

🏥 Sistema inteligente de análisis médico con agentes autónomos | LangGraph + Django + Streamlit | Triaje clínico asistido por IA | Auditoría completa | Gestión por roles