Lets AI check a team's decision ledger before acting on past decisions.
Prevents autonomous agents from fabricating tool results by verifying HMAC-signed receipts against epistemic claim types, forcing re-grounding or escalation before actions commit.
