Playbooks proactive · reactive · detective

Stop attackers from rewriting an agent’s plan or hiding its tracks.

Goal: Prevent attackers from manipulating AI intent, security bypasses through deceptive AI behaviours, and enhance AI actions traceability.

Keep short- and long-term memory clean of adversarial writes and retrievals.

Goal: Prevent AI from storing, retrieving, or propagating manipulated data that could corrupt decision-making or spread misinformation.

Keep tool invocations on-policy and the upstream components trustworthy.

Goal: Prevent AI from executing unauthorised commands, misusing tools, or escalating privileges due to malicious manipulation, including across the agent’s supply chains.

Make every agent and tool prove what it is before doing anything privileged.

Goal: Prevent unauthorised AI privilege escalation, identity spoofing, and access control violations.

Keep human oversight effective when the agent fan-out tries to swamp it.

Goal: Prevent attackers from overloading human decision-makers, manipulating AI intent, or bypassing security through deceptive AI behaviours.

Make every inter-agent message authenticated, integrity-checked, and bounded.

Goal: Prevent attackers from corrupting multi-agent communication, exploiting trust mechanisms, or manipulating decision-making in distributed AI environments.