AI agent exploited Salesforce sites; 263 objects, 55 Apex methods exposed at one portal, leading to PII and file leaks.

Anthropic's Mythos Preview was highly effective at finding vulnerability candidates, especially when analyzing source code.

Disclaimer: This article is sponsored by BulkQuant. It is intended for informational purposes only and does not constitute financial advice, investment advice, or a guarantee of trading results.

点击上方蓝字关注我，加个🌟标不迷路。 大家好，我是 cxuan，一个和 AI Agent 互相折磨的 builder。 在真实项目里，Codex 不只需要会写代码，还需要知道这个项目的规矩：用什么命令、跑什么测试、哪些文件不能动、哪些操作要先确认。 这些内容如果每次都手动敲 ...