evolve
Run one safe self-improvement cycle: hypothesize → experiment → record → (maybe) federate.
来源信息
- 仓库
- ruvnet/ruqu
- 最近来源活动
- 2026年6月17日 20:28
- 检测到的 SKILL.md 语言
- 英语
- 星标
- 19
- Forks
- 5
安装方式
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。
检查来源文件
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。
正在显示 SKILL.md
SKILL.md
来源说明 · 只读预览- name
- evolve
- description
- Run one safe self-improvement cycle: hypothesize → experiment → record → (maybe) federate.
# evolve
Run one evolution cycle.
1. Hypothesizer reads the evolution log and proposes a falsifiable change with a metric.
2. Experimenter tests it in a sandbox and records a signed kept/killed result.
3. Federator shares it to peers only if witness-signed and reproduced.
Guard against Goodharting the metric. See ADR-014 (self-evolution + federation).
在 GitHub 查看