# Elvis Saravia 转述 Rahul 的 AI 编码智能体 10 项原则

- 来源：elvis (@omarsar0)
- 发布时间：2026-06-18 00:28
- AIHOT 分数：60
- AIHOT 链接：https://aihot.virxact.com/items/cmqiayfav078gslf0pyx3360f
- 原文链接：https://x.com/omarsar0/status/2067283235759935648

## AI 摘要

Rahul 指出，fable+ 类模型本质是英语→代码解释器，Fable 5 最差。diff 大小按风险管控：高风险区（身份/数据/网络/资金）用小 diff，可经验验证代码用大 diff。软件交付速度取决于审查/合并能力而非 PR 生成，瓶颈在 lint、测试、CI、影子验证。智能体需深度理解全栈，风险优先级为安全>正确性>性能。复杂性成本变化，可能值得多维护 50% 代码换 5% 性能提升。低风险时将代码块作黑盒仅做经验验证。逻辑逐行审查成本高，只用于关键处。更快迭代需借助权限 opt-in、影子模式等护栏。

## 正文

Highly-recommended reading！

After using /loops &amp； /goal throughout my projects， I believe that verifiers and robust guardrails are imperative to get current/future coding agents to work right. You can't just YOLO your way with blind autonomous loops. It doesn't work！

### 引用推文

> rahul：1. as a mental model it is more correct to think of fable+ class models as english -> code interpreters - converts your idea into code into "correct" code regar...