自动从 Claude Code 会话中提取可重用模式，并将其保存为 learned skills 供将来使用。
npx claudepluginhub aaione/everything-claude-code-zhThis skill uses the workspace's default tool permissions.
自动在结束时评估 Claude Code 会话，以提取可重用模式并将其保存为 learned skills。
Automatically evaluates Claude Code sessions on completion to extract reusable patterns like error resolutions, workarounds, and debugging techniques, saving them as learned skills.
Automatically evaluates Claude Code sessions via Stop hook to extract reusable patterns like error resolutions, debugging techniques, and workarounds, saving them as learned skills.
Automatically extracts reusable patterns like error resolutions and workarounds from Claude Code sessions via Stop hook and saves as learned skills.
Share bugs, ideas, or general feedback.
自动在结束时评估 Claude Code 会话，以提取可重用模式并将其保存为 learned skills。
此技能作为 Stop hook 在每个会话结束时运行：
~/.claude/skills/learned/编辑 config.json 进行自定义：
{
"min_session_length": 10,
"extraction_threshold": "medium",
"auto_approve": false,
"learned_skills_path": "~/.claude/skills/learned/",
"patterns_to_detect": [
"error_resolution",
"user_corrections",
"workarounds",
"debugging_techniques",
"project_specific"
],
"ignore_patterns": [
"simple_typos",
"one_time_fixes",
"external_api_issues"
]
}
| Pattern | Description |
|---|---|
error_resolution | 特定错误是如何解决的 |
user_corrections | 来自用户纠正的模式 |
workarounds | 框架/库怪癖的解决方案 |
debugging_techniques | 有效的调试方法 |
project_specific | 项目特定约定 |
添加到你的 ~/.claude/settings.json:
{
"hooks": {
"Stop": [{
"matcher": "*",
"hooks": [{
"type": "command",
"command": "~/.claude/skills/continuous-learning/evaluate-session.sh"
}]
}]
}
}
/learn 命令 - 会话中手动提取模式Homunculus v2 采取了更复杂的方法：
| Feature | Our Approach | Homunculus v2 |
|---|---|---|
| 观察 (Observation) | Stop hook (会话结束) | PreToolUse/PostToolUse hooks (100% 可靠) |
| 分析 (Analysis) | 主上下文 | 后台 agent (Haiku) |
| 粒度 (Granularity) | 完整技能 | 原子 "instincts" (本能) |
| 置信度 (Confidence) | 无 | 0.3-0.9 加权 |
| 演进 (Evolution) | 直接从 skill | Instincts → cluster → skill/command/agent |
| 共享 (Sharing) | 无 | 导出/导入 instincts |
来自 homunculus 的关键见解:
"v1 依靠技能进行观察。技能是概率性的——它们大约 50-80% 的时间触发。v2 使用 hooks 进行观察（100% 可靠）并将 instincts 作为学习行为的原子单位。"
参见: /Users/affoon/Documents/tasks/12-continuous-learning-v2.md 获取完整规范。