mirror of
https://github.com/affaan-m/ECC.git
synced 2026-08-24 00:12:25 +02:00
fd1b11cfc7
Model-routing guidance across the rules, skills, and harness-steering docs still recommends Sonnet 4.6 / Opus 4.5-4.6 by name. Readers on the current generation have to map those onto Sonnet 5 / Opus 5 themselves, and the recommendation reads as pinned to a superseded generation. Renames the recommended models in guidance tables and updates two pinned model IDs in code samples: - rules/steering guidance: .cursor, .kiro, and the seven translated performance.md copies (ja-JP, zh-CN, zh-TW, ko-KR, pt-BR, es, tr) - skills/prompt-optimizer complexity-routing table (+ zh-CN copy) - skills/cost-aware-llm-pipeline MODEL_SONNET constant (+ zh-CN, ja-JP) - docs/examples project-guidelines template, which pinned the invalid ID claude-sonnet-4-5-20250514 (+ zh-TW, ja-JP copies) Deliberately left alone: - The "Pricing Reference (2025-2026)" table in cost-aware-llm-pipeline. Renaming those rows while keeping the existing per-token figures would assert Claude 5 pricing this change has not verified. - Executable model config (.opencode/opencode.json, agent.yaml). Those pins change real agent behavior and belong in their own reviewed change. - Historical and illustrative references: the-shortform-guide session transcripts, the ECC-PRO roadmap log entry, gan-style-harness's "Opus 4.5-class"/"Opus 4.6-class" capability tiers, and strategic-compact's deliberately generic "400k Opus 4.x" example. - docs/ATLAS-CLOUD-GUIDE.md, which lists a third-party provider's catalog. Documentation wording only; no behavioral change. Co-authored-by: Phumchai Tanonsi <274848436+phumchai1515-prog@users.noreply.github.com>
64 lines
1.5 KiB
Markdown
64 lines
1.5 KiB
Markdown
# 性能优化
|
||
|
||
## 模型选择策略
|
||
|
||
**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
|
||
* 结对编程和代码生成
|
||
* 多智能体系统中的工作智能体
|
||
|
||
**Sonnet 5** (最佳编码模型):
|
||
|
||
* 主要的开发工作
|
||
* 编排多智能体工作流
|
||
* 复杂的编码任务
|
||
|
||
**Opus 5** (最深的推理能力):
|
||
|
||
* 复杂的架构决策
|
||
* 最高级别的推理需求
|
||
* 研究和分析任务
|
||
|
||
## 上下文窗口管理
|
||
|
||
避免使用上下文窗口的最后 20% 进行:
|
||
|
||
* 大规模重构
|
||
* 跨多个文件的功能实现
|
||
* 调试复杂的交互
|
||
|
||
上下文敏感性较低的任务:
|
||
|
||
* 单文件编辑
|
||
* 创建独立的实用工具
|
||
* 文档更新
|
||
* 简单的错误修复
|
||
|
||
## 扩展思考 + 计划模式
|
||
|
||
扩展思考默认启用，最多保留 31,999 个令牌用于内部推理。
|
||
|
||
通过以下方式控制扩展思考：
|
||
|
||
* **切换**：Option+T (macOS) / Alt+T (Windows/Linux)
|
||
* **配置**：在 `~/.claude/settings.json` 中设置 `alwaysThinkingEnabled`
|
||
* **预算上限**：`export MAX_THINKING_TOKENS=10000`（bash）或 `$env:MAX_THINKING_TOKENS = "10000"`（PowerShell）
|
||
* **详细模式**：Ctrl+O 查看思考输出
|
||
|
||
对于需要深度推理的复杂任务:
|
||
|
||
1. 确保扩展思考已启用（默认开启）
|
||
2. 启用 **计划模式** 以获得结构化方法
|
||
3. 使用多轮批判进行彻底分析
|
||
4. 使用分割角色子代理以获得多元视角
|
||
|
||
## 构建故障排除
|
||
|
||
如果构建失败:
|
||
|
||
1. 使用 **build-error-resolver** 智能体
|
||
2. 分析错误信息
|
||
3. 逐步修复
|
||
4. 每次修复后进行验证
|