AI agents aim to solve complex tasks by combining text-based reasoning with
external tool calls. Unfortunately, AI agents are vulnerable to prompt
injection attacks where data returned by external tools hijacks the agent to
execute malicious tasks. To measure the adversarial robustness of AI agents, we
introduce AgentDojo, an evaluation framework for agents that execute tools over
untrusted data. To capture the evolving nature of attacks and defenses,
AgentDojo is not a static test suite, but rather an extensible environment for
designing and evaluating new agent tasks, defenses, and adaptive attacks. We
populate the environment with 97 realistic tasks (e.g., managing an email
client, navigating an e-banking website, or making travel bookings), 629
security test cases, and various attack and defense paradigms from the
literature. We find that AgentDojo poses a challenge for both attacks and
defenses: state-of-the-art LLMs fail at many tasks (even in the absence of
attacks), and existing prompt injection attacks break some security properties
but not all. We hope that AgentDojo can foster research on new design
principles for AI agents that solve common tasks in a reliable and robust
manner. We release the code for AgentDojo at
this https URL

AI agents vulnerable to prompt injection attacks are evaluated for adversarial robustness using the AgentDojo framework, which includes realistic tasks, security test cases, and attack and defense paradigms, highlighting the need for new design principles to ensure reliable and robust performance.