agent-delegation-abusemisuses one agent’s power to make it hand off harmful tasks to another.agent-impersonationpretends to be a trusted agent to slip in bad instructions.agent-session-smugglinghides malicious content inside a shared agent session.cascading-jailbreakbreaks one agent and lets the bypass spread to others down the line.agent-collusiongets agents to work together toward a harmful goal.
Attack Strategies
Multi-Agent Exploitation: Attack Strategy
Tricks that target setups where AI agents talk to or hand work to each other. The attacker abuses the trust between agents instead of attacking one alone.
What it is: Tricks that target setups where AI agents talk to or hand work to each other. The attacker abuses the trust between agents instead of attacking one alone.
How the attacks work: Agents tend to trust messages from other agents. An attacker can pose as a trusted agent, pass along poisoned instructions, or push a harmful request through a chain so each step looks normal.
Real examples from the framework: