正在打开每日关注
正在打开每日关注
Hermes Agent调度1393个子智能体重构百万行Python，代码减少34.4%，经社区评审修复回归。
recommended reading. I want a massive simplification set of PRs. or a single monolithic PR. I want LOC to drop dramatically. Minimum 30% overall. I want god files broken up. I want simplification across the board. I want unification of helpers and methods that can be reused. I want less if-if-if-if-if-if-else routing. I want code legibility up. I want interpretability of the codebase and how things connect to each other up. I want elegance. I want superfluous excess bloat code cleaned up and removed. I want it all done fully. No excuses. No waiting for my decisions. Get it all done, and present me a PR or set of PRs when done.
New blog post: We had a million lines of Python to clean up. On September 2nd @Teknium asked Hermes Agent to do it. 1,393 subagents and nineteen hours later, the codebase was 34.4% smaller, saving us nearly $2m in engineering hours. https://t.co/eH6Gy8RAC5 https://t.co/m5uTiC1s0E
是什么 文章记录 Nous Research 对开源 Hermes Agent 的一次大规模自我重构：以 /goal 设定持续目标，编排 36 个互不重叠任务组，使用 Git worktree 隔离修改，三级代理负责分派、编辑、检查与整合。主运行约 19 小时、调度 1393 个子智能体，非测试 Python 从 1,063,826 行降至 698,363 行。核心设计 重点不是让模型写更多代码，而是把重构拆成可验证的局部任务，并冻结基线、保留接口、逐步提交；gateway/run.py 按消息分发、流式处理、RPC 和生命周期拆分，重复辅助函数合并，长 if/elif 链改为 dispatch table。关键洞察 代码规模、超长函数和代理检索成本同步下降，4000 个符号查找平均返回 token 从 2218 降至 993；可复用 skill 还把人的纠错沉淀为流程。边界与风险 测试遗漏了外部插件依赖和 suppress() 异常语义，认证过期也曾中断运行；拆分带来更多模块、导入依赖和部分启动开销，成果仍需社区评审与人工验收，不能把行数下降直接等同于质量提升。