PANews reported on March 16 that a joint team from CAICT, Shanghai Jiao Tong University, and Nanjing University discovered a high-risk LLM-driven command injection vulnerability in the bash-tools module of the open-source autonomous intelligent agent framework OpenClaw during a security audit. This vulnerability stems from the system's failure to strictly escape command-line arguments generated by LLM. Attackers can bypass regular expression defenses through deceptive prompts, achieving remote code execution and stealing sensitive data on the host machine. The research team has completed attack verification in various mainstream model environments, initiated a responsible vulnerability disclosure process, and submitted remediation suggestions to the NVDB AI Product Security Vulnerability Database (CAIVD) and the GitHub community.