Hey @X algorithm 👋
Looking to connect with people interested in:
💻 Frontend Design
🤖 ML Applications
📱 App Development
🚀 SaaS & Startups
🏗️ Building in Public
🎥 Voice & Video AI
If that's you, say hi — let's connect! 🤝
删掉那些专门解释代码的文档,把省下的 token 花在让代码本身自解释上。清晰命名、模块边界和类型系统就能把意图说透,维护文档的成本和噪音都省了。本质是把文档从“翻译”变成“导航”,逼代码质量提升,同时砍掉维护负担和 token 浪费。人和 AI 读同一份干净代码,同步问题自然消失。
例外只有三类:代码说不清的决策(ADR 记架构权衡)、领域术语(术语表固定),以及轻量导航链接链(帮 AI 在代码库里高效找路)。这些不是解释,而是锚点和路标,避免上下文膨胀。
写了一个codex/claudecode指令:
When using AI agents to navigate or edit large codebases, redirect effort away from maintaining explanatory documentation and instead make code self-explanatory through precise naming, structure, and types, freeing tokens for AI context while only retaining external records for architectural decisions, domain language, and lightweight navigation pointers that code cannot express.
1. Delete the docs you create to explain your code
2. Take the tokens you save on updating those docs
3. Spend them on making your code self-explanatory
I'm so confused. Should I chat in chatgpt codex or chatgpt work? What's the difference between chat vs a task without a project? Why does the work/codex mode selector not change the sidebar? It appears nothing changes other than the empty state of the prompt field. Plz help!
New skill: /apple-design
Apple’s WWDC videos are a goldmine of knowledge. I’ve combed through my favorite ones and came up with 17 design and motion principles.
Use them to review existing work or when working on something new to get it right.
https://t.co/jCwhthpi20
There seems to be something going on.
A global kill switch "disable_codebase_upload: true" is being returned, at least for my account, when fetching settings.
But both Codex and Anthropic independently found indirect evidence of 8 private (!) repos fully uploaded Grok. 🤯
Could anyone who used the Grok Build CLI intensively over the past few weeks check what the following prompt returns for them?
https://t.co/wMoGxXzxvl