TrustGate
Define, test, and enforce AI usage policies
TrustGate is the enterprise control layer for what AI is allowed to do. Author policies (no secrets in prompts, PII stays on approved providers, approved models only, human-approval workflows, RAG citation requirements, token/cost ceilings, memory-write and tool-usage rules, logging) and test any proposed AI action against them.
The decision engine is deterministic, not an LLM guess: each rule matches and the strongest enforcement wins — monitor, warn, require approval, or block — with the exact reasons and a recommended fix. Violations are logged, approvals queue for a human, and everything is auditable.
A Pro workspace gated behind Nyquest Pro, enforced on both the frontend and the backend. Export a portable Nyquest Policy Package and enforce it across the Foundry.
Live now. Requires a Nyquest Pro subscription. Deterministic policy decisions with reasons — not vibes.