AiNIR v1.0 RC Candidate

AiNIR v1.0 RC Candidate is a bounded public demo of a semantic trust layer for inspecting AI-generated workflow semantics before lowering or execution.

This release candidate demonstrates:

• strict draft intake
• registry and evidence checks
• Trust Gate decisions and TrustReceipt replay
• negative conformance and golden trace coverage
• lowering eligibility for the safe outbox example

Boundary:

• not a v1.0 final release
• not a production runtime
• not a verifier for arbitrary AI-generated code
• private archive material and extended policy packs are not included