菜单
Reproduce a generation: copy each winner's whole agent directory into the next generation and wipe its session memory (children inherit structure, not history). Use for evolving agents/teams across generations. `evolve()`.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/sancovp/cave-teams --skill cave-evolve命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
基于 SOC 职业分类
| name | cave-evolve |
| description | Reproduce a generation: copy each winner's whole agent directory into the next generation and wipe its session memory (children inherit structure, not history). Use for evolving agents/teams across generations. `evolve()`. |
The genetic operator over agent directories: evolve(winner_dirs, out_dir) copies each winner's whole AIOS dir into out_dir/{prefix}_{i} and removes its knowledge of previous sessions — the next generation, born advanced (the dir) but memory-free. Returns the child dirs.
from cave_teams import evolve, select_winners
children = evolve(winner_dirs, out_dir="gen2") # -> list of child dirs
Not a cave() build-op — evolve is a genetic filesystem operation (it returns child directories, not a Link). Call it directly. It pairs with cave-tournament (select winners) and cave-sim (run it over generations).
cave-tournament · cave-sim · cave-world
Part of the cave-teams plugin — the language + full DSL is the cave-teams skill; drive any pattern from data with the cave skill.