Claude CN Blog
Welcome to the Claude CN Blog! Explore Claude AI’s latest technology, share development practice experiences, providing Chinese developers with authoritative technical guides and ecosystem observations.
🎯 Content Categories
💬 Subscribe & Engage
📚 Latest Articles
Browse articles below for the latest Claude AI technical updates and development practice guidance.
Anthropic Official Skills Repository: 17 Skills — What to Install, How, and Why
A breakdown of the current 17 top-level Skills directories in Anthropic's official repository. Categorized by creative design, document processing, development tools, enterprise collaboration, and the meta skill — with installation commands, mechanism analysis, and practical use cases.
April 5, 2026
'It Feels Like Cheating': What Actually Matters in the Age of AI Coding
Using LLMs to code 'feels like cheating.' When code itself is no longer scarce, the value hierarchy — 10X engineers, specialists vs. generalists, and the desire to build — is being rewritten.
February 7, 2026
Claude Code for Designers: The Shortest Workflow from Idea to a Shippable Demo
A complete designer's guide to shipping real products with Claude Code — from installation and terminal basics to GitHub, Vercel deployment, Supabase auth, OpenAI integration, and prompt templates.
February 6, 2026
Claude Skills: Professional Intro — Turn "Using AI" into Repeatable Engineering Capability
An engineering-focused intro to Claude Skills: what Skills are, how progressive disclosure works, how Skills divide work with Prompts/Projects/Subagents/MCP, and a practical checklist to ship your first reusable Skill.
January 24, 2026
Claude Code in China: Install, Third-party Providers, and Multi-model Switching (Windows/macOS)
Get Claude Code running locally: from Node/CLI installation, configuring Anthropic or Anthropic-compatible third-party providers, to scripting one-click model/provider switches and syncing with VS Code.
January 23, 2026
Claude Code Best Practices: Turn Context into Team Productivity
A team-executable playbook: verification loops, Plan Mode, CLAUDE.md, permissions/sandboxing, and operationalizing Hooks/MCP/Plugins/Subagents/Skills.
January 22, 2026
Claude Skills Architecture: Progressive Disclosure, Runtime, and Sandboxing
An architectural view of Claude Skills: why context engineering matters, how progressive disclosure controls token cost, how on-demand scripts/resources improve determinism, and how runtime/sandbox/permissions enable governance.
January 14, 2026
Claude Cowork System Prompt: How Anthropic Builds a Digital Coworker
A guided reading of Claude Cowork mode system prompts: product positioning, safety boundaries, interaction philosophy, tool usage, knowledge limits, citations, and artifacts.
January 14, 2026
Claude Cowork System Prompt (Full Analysis): How Anthropic Engineers a Digital Coworker
A deeper analysis of Claude Cowork system prompts: product architecture and matrix, safety boundaries, interaction philosophy, wellbeing, neutrality, tools, computer-use execution, honesty about limits, citation rules, and artifacts.
January 14, 2026
Claude Cowork Landing Guide: Task Templates, Security Guardrails & Artifact Delivery
Focusing on practical deployment of Claude Cowork: task lifecycle, Artifacts delivery, least privilege, audit/rollback governance, with ready-to-use task description templates and pilot checklists.
January 14, 2026