**Opus 5或已解决AI代理浏览器提示注入难题** _Opus 5 may have solved browser-based prompt inject
原帖
**Opus 5或已解决AI代理浏览器提示注入难题**
_Opus 5 may have solved browser-based prompt injection, the biggest security flaw haunting AI agents_
> Anthropic的Opus 5结合Auto Mode在129项测试中实现零提示注入成功率,显著优于无防护时的3.7%,有望攻克AI代理面临的最大安全漏洞。
**来源信息**
- **来源**:The Decoder:AI News(RSS)
- **分类**:ai-models
- **发布时间**:2026-07-25 18:43(北京时间)
- **原文**:[打开原文](https://the-decoder.com/opus-5-may-have-solved-browser-based-prompt-injection-the-biggest-security-flaw-haunting-ai-agents)