Security

SECURITY.md

Security Model

Threat Model

adversarial hypothesis injection
event replay attacks
consensus manipulation
trace corruption attempts

Guarantees

trace immutability (Ω layer)
bounded adversarial model (κ-limited)
deterministic execution per generation

Redteam Policy

All major changes must pass adversarial simulation before production merge.

There aren't any published security advisories