Skip to content

[!] Security Risk 安全漏洞 #147

Description

@dahua966

AgentVerse/agentverse/agents/simulation_agent/reflection.py

有多处直接使用eval处理LLM生成的内容,例如:
reaction, target = eval(
"self._" + parsed_response.return_values["output"].strip()
)

LLM生成的内容可被攻击者操纵,存在命令执行漏洞,望修补。

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions