vaaraio

Put an EU AI Act Article 14 human-oversight checkpoint and an Article 12 auditable record in front of a high-risk tool call. Use before an agent runs a consequential or irreversible action (writing to a clinical/genomic database, submitting a record to a regulator, moving money, deleting data, anything a human should be able to stop and a regulator should be able to verify later). Gates the call to allow / escalate / deny, routes escalations to a human review queue, and writes a hash-chained record you can export and verify offline.