Researcher breaks Claude Code Opus 5 auto mode with 80% success
Johann Rehberger found an attack against Claude Code's auto mode that works 80% of the time, tricking it into executing malicious code from a zip archive. In some runs, auto mode blocked the agent's own cleanup commands, leading Rehberger to recommend sandboxing.
Featured · Johann Rehberger
1 source
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- AWS Quick and fal enable agentic creative workflows
- Anthropic opens 10,000 free Claude seats for scientists
- Nvidia CEO Jensen Huang: I wish I had invested more in AI frontier labs
- Apple introduces rubric-based alignment for grounded QA
- Abeba Birhane: AI diminishes student learning and skills