Harness 工程：AI 自我改进的关键设计模式
原标题：Harness Engineering for Self-Improvement
AI 摘要
Lil'Log 发布文章《Harness Engineering for Self-Improvement》，探讨了递归自我改进（RSI）中“harness”（即围绕基础模型的系统）的重要性。文章指出，harness 设计模式包括工作流自动化、文件系统作为持久记忆、子代理与后台任务，并以编码代理（如 Claude Code、Codex）为例说明其核心接口。该文强调 harness 工程对 AI 部署和 RSI 的关键作用，并总结了相关研究趋势。
正文节选
The concept of recursive self-improvement (RSI) dates back to I. J. Good (1965), where he defined an “ultraintelligent machine” as a system that can surpass humans in all intellectual activities and design better machines to improve itself. Yudkowsky (2008) used the phrase “recursive self-improvement” for a specific feedback loop: an AI uses its current intelligence to improve the cognitive machinery that produces its intelligence. This feedback loop in modern AI may indicate the model rewriting