Obiguard's Policy engine lets you define exactly what each AI agent is allowed to do — and block or flag anything outside that envelope in real time.
Criteria bind a detector (or custom rule) to a direction — prompt, response, or tool-call — and an action. One criterion, one decision point.
Bundle related criteria into a named set. A support-agent set might include no-pii-in-responses, no-competitor-mention, and safe-tool-list.
Link a Policy Set to one or many AI Agents. The set travels with the agent — across models, frameworks, and deployment environments.
Any criterion can route violations to the Review Queue for human judgment. Every decision is written back to the Audit Ledger.