Developer-first security layer for AI applications. Deterministic detection of prompt injection across 13 attack categories.
GuardWeave: risk-adaptive prompt injection defense for hosted APIs and local LLMs.
