Blocking at the Boundary: Auditing Long-Horizon Agents against Staged Prompt Injection
An automated, feedback-guided attack generation pipeline is built and applied to Claude Code and Codex in their native runtimes, showing that production agents are vulnerable to context-aware, multi-step injection over long horizons.
Jing-Kai Liu, Yu-Fei Han, Xiaoting Lyu et al.
· 0 citations