Make Agent Defeat Agent: Automatic Detection of Taint-Style Vulnerabilities in LLM-based Agents

Latest version: https://github.com/LFYSec/AgentFuzz

The repository includes the dataset and prototype source code used in the AgentFuzz experiments. Specifically, `dataset.md` describes the addresses and versions of the applications in the dataset, while `readme.md` provides instructions on how to set up the prototype and start fuzzing.