docs: 完成所有文档的中文翻译并应用到项目

This commit is contained in:
xuxiang
2026-01-28 00:12:54 +08:00
parent 0ced59a26b
commit e133f58e1c
76 changed files with 6808 additions and 6170 deletions

View File

@@ -1,47 +1,47 @@
# Performance Optimization
# 性能优化(Performance Optimization
## Model Selection Strategy
## 模型选择策略(Model Selection Strategy
**Haiku 4.5** (90% of Sonnet capability, 3x cost savings):
- Lightweight agents with frequent invocation
- Pair programming and code generation
- Worker agents in multi-agent systems
**Haiku 4.5**(具备 Sonnet 90% 的能力,节省 3 倍成本):
- 频繁调用的轻量级智能体Agents
- 结对编程与代码生成
- 多智能体系统中的执行者智能体Worker agents
**Sonnet 4.5** (Best coding model):
- Main development work
- Orchestrating multi-agent workflows
- Complex coding tasks
**Sonnet 4.5**(最佳编程模型):
- 主力开发工作
- 编排多智能体工作流Workflow
- 复杂的编程任务
**Opus 4.5** (Deepest reasoning):
- Complex architectural decisions
- Maximum reasoning requirements
- Research and analysis tasks
**Opus 4.5**(最深层的推理能力):
- 复杂的架构决策
- 极高的推理需求
- 研究与分析任务
## Context Window Management
## 上下文窗口管理(Context Window Management
Avoid last 20% of context window for:
- Large-scale refactoring
- Feature implementation spanning multiple files
- Debugging complex interactions
在以下场景中避免触及上下文窗口Context Window)最后 20% 的容量:
- 大规模重构
- 涉及多个文件的功能实现
- 调试复杂的交互逻辑
Lower context sensitivity tasks:
- Single-file edits
- Independent utility creation
- Documentation updates
- Simple bug fixes
对上下文敏感度较低的任务:
- 单文件编辑
- 独立工具函数创建
- 文档更新
- 简单的 Bug 修复
## Ultrathink + Plan Mode
## Ultrathink + 计划模式(Plan Mode
For complex tasks requiring deep reasoning:
1. Use `ultrathink` for enhanced thinking
2. Enable **Plan Mode** for structured approach
3. "Rev the engine" with multiple critique rounds
4. Use split role sub-agents for diverse analysis
对于需要深度推理的复杂任务:
1. 使用 `ultrathink` 以获得增强的思维过程
2. 启用 **计划模式(Plan Mode** 以采用结构化方法
3. 通过多轮评审Critique rounds)来“预热引擎”
4. 使用分角色的子智能体Sub-agents进行多样化分析
## Build Troubleshooting
## 构建故障排除(Build Troubleshooting
If build fails:
1. Use **build-error-resolver** agent
2. Analyze error messages
3. Fix incrementally
4. Verify after each fix
如果构建失败:
1. 使用 **build-error-resolver** 智能体
2. 分析错误消息
3. 采用增量方式修复
4. 每次修复后进行验证