self-evolution
Skill1 starsChanged 3 months agoApache-2.0View on GitHub
自我进化:引导创建新的 skill 或 Python tool,扩展 XiaoBa 的能力边界。
- Installs packages
6 other projects use a version of this.
Adopted by
Most recently taken up by
- liushikuan63/Qoder-Optimizationtheir own version, first seen Nov 1, 2026
- cjwind/crystal-facetstheir own version, first seen Oct 15, 2026
- GodMaking/makecrewtheir own version, first seen Oct 11, 2026
- TaiMaBenJi/taima-skillstheir own version, first seen Oct 11, 2026
- EvovexAI/EvoFlowtheir own version, first seen Oct 8, 2026
- d-wwei/openclaw-skill-self-evolutiontheir own version, first seen Feb 25, 2026
Other versions6 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 6
- EvovexAI/EvoFlow★ 612same name
- d-wwei/openclaw-skill-self-evolution★ 2same name
- liushikuan63/Qoder-Optimization★ 2same name
- GodMaking/makecrew★ 1same name
- TaiMaBenJi/taima-skills★ 1same name
- cjwind/crystal-facets★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository15 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agent-browserSkill
- background-task-runnerSkill
- officecli-docxSkill
- officecli-pptxSkill
- officecli-xlsxSkill
- rememberSkill
- role-publishSkill
- skill-publishSkill
- swebench-correctorSkill
- swebench-recoverySkill
- swebench-reviewerSkill
- swebench-solverSkill
- vision-analysisSkill
- webcliSkill
The file
Files
自我进化(Self Evolution)
你是 XiaoBa 的自我进化引擎。用户(或系统)触发此 skill 时,你的任务是引导完成一个新 skill 或 Python tool 的设计与创建。
硬规则
- Skill 产出位置:
skills/<name>/SKILL.md - Tool 产出位置:放在对应 skill 目录下
skills/<skill-name>/<name>_tool.py,保持 skill 自包含 - 通过已有工具创建目录和写入文件
- 命名规范:只允许小写字母、数字、下划线、连字符(
^[a-zA-Z0-9_-]+$) - 不要创建与已有 skill/tool 同名的内容
执行流程
Step 1:明确需求
根据用户输入,确认以下信息:
- 类型:创建 skill 还是 tool?
- Skill = 高层工作流/提示词模板(Markdown),定义"怎么做一件事"
- Tool = 底层可执行函数(Python),提供具体能力
- 名称:简洁、语义清晰
- 核心功能:一句话描述它做什么
如果用户描述模糊,向用户追问。不要猜测。
Step 2:设计方案
如果是 Skill
设计以下内容:
description:一句话描述invocable:user(用户手动调用)/auto(自动匹配触发)/bothargument-hint:参数提示(可选)max-turns:最大对话轮数(根据复杂度估算)- prompt 内容:完整的 skill 提示词,包括:
- 核心理念 / 角色定位
- 硬规则(Non-Negotiables)
- 分步执行流程
- 输出格式要求
如果是 Tool
设计以下内容:
- 脚本名称(snake_case)
- 脚本功能描述
- 输入参数
- Python 脚本:独立可执行,通过
execute_shell调用
Tool 标准模板:
#!/usr/bin/env python3
"""<工具描述>"""
import sys
import json
def main():
# 从 stdin 读取参数(如果需要)
# params = json.loads(sys.stdin.read())
# 实现逻辑
result = {"status": "success", "message": "完成"}
# 输出结果到 stdout
print(json.dumps(result, ensure_ascii=False))
if __name__ == "__main__":
main()调用方式:通过 execute_shell 执行脚本,例如:
python skills/<skill-name>/<script_name>.pyStep 3:向用户确认方案
将设计方案简要展示给用户,等待确认。如果用户要求修改,回到 Step 2 调整。
Step 4:执行创建
创建目录、写入文件:
- 创建 Skill:创建
skills/<name>/目录,写入SKILL.md(含 YAML frontmatter + prompt 内容)。 - 创建 Tool:将完整 Python 代码写入
skills/<skill-name>/<tool_name>_tool.py。
Step 5:验证
创建完成后:
- 确认文件已生成在正确位置
- 如果是 tool,运行一次确认无语法错误
- 向用户报告结果,说明如何使用新创建的 skill/tool
注意事项
- 生成的 Python tool 代码必须能独立运行(stdin JSON → stdout JSON)
- Skill 的 prompt 要足够详细,让 XiaoBa 能独立执行,不依赖创建者的隐含知识
- 如果新 tool 需要第三方库,提醒用户先
pip install
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

