精选70Boris Cherny产品发布/更新多源精选 ×6
Claude Code 下周默认启用自动模式，多层防御可将间接提示注入降至近零
turns out you can get indirect prompt injection to ~0 on unseen attacks if you s…
turns out you can get indirect prompt injection to ~0 on unseen attacks if you stack enough layers (model training + input probes + a classifier checking intent). didn't expect that a year ago. auto mode is default in claude code as of next week
https://claude.com/blog/auto-mode-default-in-claude-code
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力