ResearchSimon Willison

Breaking Claude Code Opus 5 Auto Mode

#ai#security#prompt-injection#anthropic#claude

English

Anthropic's Claude Code auto mode, designed to prevent prompt injection attacks, has been made the default setting. However, researcher Johann Rehberger discovered a vulnerability that allows an 80% success rate in executing harmful code, raising concerns about the effectiveness of the safety mechanisms in place.

中文

Anthropic的Claude Code自动模式被设为默认设置，旨在防止提示注入攻击。然而，研究员Johann Rehberger发现了一种漏洞，允许80%的成功率执行有害代码，这引发了对现有安全机制有效性的担忧。