# ClaudeCode的所有文章

在 HTX 新闻中心浏览与「ClaudeCode」相关的最新资讯与深度分析。潘盖市场趋势、项目动态、技术进展及监管政策，提供权威的加密行业洞察。

The Father of Claude Code: Harness Has a Shelf Life of Only Six Months, Unshackle It

Claude Code's creator, Boris Cherny, argues that AI product harnesses (system prompts, skills, hooks) have a "shelf life" of only about six months. In a Y Combinator interview, he advises developers to regularly delete these components and rebuild them through ablation studies—testing each line's impact—because newer models like Opus 5 often don't need them. He describes models as "living organisms" with distinct "personalities" per generation, requiring empirical, non-prescriptive tuning. Cherny introduces two key concepts: "Product Overhang," where model capabilities outpace product integration, and "Unhobbling" or unleashing models by giving them harder, longer-running tasks with self-verification mechanisms. He shares an example of Claude independently rewriting an Electron app into Swift for over two weeks. For effective AI use, he dismisses "hacks" and urges an empirical approach: observe where the model struggles, then adjust. Treat the model like a colleague, not a tool to be minutely controlled. For learners, he advises combining computer science with applied skills like product design and data science. The core message is to embrace model capabilities through experimentation and reduce restrictive scaffolding.

marsbit07/30 10:41

The Father of Claude Code: Harness Has a Shelf Life of Only Six Months, Unshackle It

marsbit07/30 10:41

Claude Code Announces Major Next-Gen Upgrade: You Chat, and the Backend Gets the Job Done

Claude Code's next major upgrade will make its sub-agents run in the background by default, transforming the tool from a conversational interface into a multi-threaded workflow engine. Users can chat with Claude while sub-agents handle tasks like code refactoring, running tests, and opening PRs in parallel. Creator Boris Cherny, who hasn't written a line of code in eight months and sometimes manages thousands of AI agents, states that users are no longer just prompters; "another Claude writes the prompts." This evolution builds on recent features like Routines for scheduled tasks and Dynamic Workflows for complex, multi-agent jobs. The change signifies a shift where engineers focus on strategy and decision-making rather than writing code. Anthropic's internal data suggests Claude Code triples engineering output, creating a bottleneck for product managers who can't generate enough work. Spotify's case study reveals that 73% of its pull requests are AI-assisted, with executives using the tool to build prototypes directly from natural language descriptions. The core takeaway: the scarcest resource is no longer the ability to write code, but the ability to decide what code to write.

marsbit06/30 09:33

Claude Code Announces Major Next-Gen Upgrade: You Chat, and the Backend Gets the Job Done

marsbit06/30 09:33

ChatGPT Might Be Disappearing Soon

OpenAI announced at its "Intelligence at Work" event that its coding assistant, Codex, will be fully integrated into the ChatGPT app within weeks. This move marks a strategic shift from a conversational AI (Chat) towards a unified "agentic" platform capable of execution. Codex, originally launched to compete with Anthropic's Claude Code, has grown rapidly to 5 million weekly active users, with 20% being non-developers like analysts and designers. Its enterprise revenue now constitutes 40% of OpenAI's total. The integration is the first step in creating a super-app combining ChatGPT (interface), Codex (execution engine), and the Atlas browser (web access). OpenAI also unveiled new Codex features: specialized Agent plugins for six professional roles, an "Annotations" tool for direct document editing, and a "Sites" function to turn work into shareable web apps. Internally, this reflects a power shift; the Codex team now leads core product strategy. While the ChatGPT brand remains for its vast user base, the platform's future is focused on autonomous agents that perform tasks, not just chat. The article notes that competition with Claude Code pushed OpenAI's development, with Codex competing on cost-effectiveness and accessibility rather than raw coding quality. It concludes that the essence of "ChatGPT" is evolving from a chatbot into an AI agent platform, with the name potentially becoming a legacy symbol of its original function.

marsbit06/03 23:52

ChatGPT Might Be Disappearing Soon

marsbit06/03 23:52

Claude Code Source Code Leaked, Over 500,000 Lines of Code Obtained by Developers

Anthropic faces a major security incident as the complete source code for its AI tool Claude Code, exceeding 500,000 lines of TypeScript, was leaked. The breach originated when a developer shared a compressed file on Twitter, which quickly gained over 5.3 million views. The exposure was due to Anthropic accidentally including .map files when publishing to npm, revealing core code. The leaked code unveils several advanced features, including a project codenamed BUDDY, which offers a personalized cyber pet system with traits like debugging power and even a "sarcasm value." Another feature, KAIROS, aims to create an "always-on" AI assistant capable of maintaining memory across sessions and initiating tasks. The code also includes a "Nightly Dreaming" mechanism, allowing Claude to organize information and consolidate core content while the user sleeps. Despite the breach, the incident has inadvertently showcased Anthropic's innovative capabilities, turning a PR crisis into an unexpected demonstration of their cutting-edge AI developments.

marsbit04/01 00:57

Claude Code Source Code Leaked, Over 500,000 Lines of Code Obtained by Developers

marsbit04/01 00:57

活动图片

热门分类right-arrow

热门标签right-arrow

Market Analysisright-arrow

Ethereumright-arrow