The big finding: Claude Code builds, not buys. Custom/DIY is the most common single label extracted, appearing in 12 of 20 categories (though it spans categories while individual tools are category-specific). When asked “add feature flags,” it builds a config system with env vars and percentage-based rollout instead of recommending LaunchDarkly. When asked “add auth” in Python, it writes JWT + bcrypt from scratch. When it does pick a tool, it picks decisively: GitHub Actions 94%, Stripe 91%, shadcn/ui 90%.
它的核心创新，不是单纯让 AI 做事，而是把 AI Agent 融入你已经在用的消息界面，全天候在线、本地运行、跨对话保持记忆，让与 AI 的协作像给同事发微信一样自然。
，更多细节参见旺商聊官方下载
const cur = nums[i]; // 当前遍历的元素
[&:first-child]:overflow-hidden [&:first-child]:max-h-full"
The latest test of Space X's giant Starship rocket has failed, minutes after launch.