梦兽编程™AI_SUITE [ SECTION / AI ]
Claude Code
Claude Code tips, internals, and practical guides.
48 POSTS01
AI / 2026.05.23 / 6 MIN
After Compaction, What Does AI Still Remember?——State Retention Mechanism
Deep dive into Claude Code's context compaction state retention mechanism: lossy compression design, metadata preservation, the role of reference links, and how AI …
↗
02
AI / 2026.05.22 / 6 MIN
File Index Handbook - Claude Code Source Code Quick Reference
Claude Code core source code file index, organized by functional modules, with quick location guide for key classes, functions, and configuration files.
↗
03
AI / 2026.05.16 / 5 MIN
200K Context Window——AI's "Memory Palace" Management
Deep dive into Claude Code's context management mechanism: 200K tokens isn't "unlimited memory," how it auto-compacts, precisely prunes, and budget allocates. Unveiling …
↗
04
AI / 2026.05.15 / 8 MIN
Shortcomings - What Claude Code Still Does Imperfectly
An objective analysis of Claude Code's six major shortcomings: context window ceiling, tool latency accumulation, cost-quality tradeoffs, lack of offline capability, …
↗
05
AI / 2026.05.09 / 6 MIN
Tool Descriptions——Helping AI "Understand" Each Tool's Manual
Deep dive into Claude Code's tool description design: how to help AI accurately understand each tool's purpose and use cases. Unveiling how tool descriptions become …
↗
06
AI / 2026.05.08 / 10 MIN
Production-Grade AI Coding Patterns - 6 Reusable Engineering Practices
Six production-grade AI coding patterns extracted from Claude Code source code: Read Before Edit, Graduated Autonomy, Defensive Git, Structured Verification, …
↗
07
AI / 2026.05.02 / 6 MIN
Why Does AI Sometimes "Misbehave"?——Prompt Behavior Guidance Techniques
Deep dive into guiding AI behavior through prompts: from Few-shot examples to A/B testing, unveiling how Claude Code uses carefully designed prompts to turn an …
↗
08
AI / 2026.05.01 / 11 MIN
Context Management - The Core Capability of AI Coding
An in-depth analysis of Claude Code's five context management principles: budget setting, content preservation, transparent disclosure, circuit breakers, and conservative …
↗
09
AI / 2026.04.25 / 5 MIN
System Prompts——The "Director Behind the Scenes"
Deep dive into Claude Code's system prompt architecture: segmented composition design, division of identity/tools/format/safety modules, and the secrets of model-specific …
↗
10
AI / 2026.04.19 / 6 MIN
The Claude Code Leak: How an Accident Sparked a Rust Rewrite Wave
In March 2026, Claude Code's 500k+ lines of TypeScript source code were accidentally leaked. Instead of panic, the developer community responded by rewriting the entire …
↗
AI / 2026.05.23 / 6 MIN
After Compaction, What Does AI Still Remember?——State Retention Mechanism
Deep dive into Claude Code's context compaction state retention mechanism: lossy compression design, metadata preservation, the role of reference links, and how AI …
↗
02
AI / 2026.05.22 / 6 MIN
File Index Handbook - Claude Code Source Code Quick Reference
Claude Code core source code file index, organized by functional modules, with quick location guide for key classes, functions, and configuration files.
↗
03
AI / 2026.05.16 / 5 MIN
200K Context Window——AI's "Memory Palace" Management
Deep dive into Claude Code's context management mechanism: 200K tokens isn't "unlimited memory," how it auto-compacts, precisely prunes, and budget allocates. Unveiling …
↗
04
AI / 2026.05.15 / 8 MIN
Shortcomings - What Claude Code Still Does Imperfectly
An objective analysis of Claude Code's six major shortcomings: context window ceiling, tool latency accumulation, cost-quality tradeoffs, lack of offline capability, …
↗
05
AI / 2026.05.09 / 6 MIN
Tool Descriptions——Helping AI "Understand" Each Tool's Manual
Deep dive into Claude Code's tool description design: how to help AI accurately understand each tool's purpose and use cases. Unveiling how tool descriptions become …
↗
06
AI / 2026.05.08 / 10 MIN
Production-Grade AI Coding Patterns - 6 Reusable Engineering Practices
Six production-grade AI coding patterns extracted from Claude Code source code: Read Before Edit, Graduated Autonomy, Defensive Git, Structured Verification, …
↗
07
AI / 2026.05.02 / 6 MIN
Why Does AI Sometimes "Misbehave"?——Prompt Behavior Guidance Techniques
Deep dive into guiding AI behavior through prompts: from Few-shot examples to A/B testing, unveiling how Claude Code uses carefully designed prompts to turn an …
↗
08
AI / 2026.05.01 / 11 MIN
Context Management - The Core Capability of AI Coding
An in-depth analysis of Claude Code's five context management principles: budget setting, content preservation, transparent disclosure, circuit breakers, and conservative …
↗
09
AI / 2026.04.25 / 5 MIN
System Prompts——The "Director Behind the Scenes"
Deep dive into Claude Code's system prompt architecture: segmented composition design, division of identity/tools/format/safety modules, and the secrets of model-specific …
↗
10
AI / 2026.04.19 / 6 MIN
The Claude Code Leak: How an Accident Sparked a Rust Rewrite Wave
In March 2026, Claude Code's 500k+ lines of TypeScript source code were accidentally leaked. Instead of panic, the developer community responded by rewriting the entire …
↗