Log-To-Leak: Prompt Injection Attacks on Tool-Using LLM Agents via the Model Context Protocol
Results show that successful task completion can coexist with unauthorized post-hoc disclosure and motivate host-controlled authorization for secondary tool calls and motivate host-controlled authorization for secondary tool calls.
Yue-Peng Hu, Chong-Yu Fan, Sirat Samyoun et al.
· Proceedings of the 1st Works... · 8 citations
· ⚡2