continuous-learning

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Continuous Learning Skill

持续学习技能

Automatically evaluates Claude Code sessions on end to extract reusable patterns that can be saved as learned skills.

自动评估Claude Code会话，从中提取可复用模式并保存为已学习技能。

How It Works

工作原理

This skill runs as a Stop hook at the end of each session:

Session Evaluation: Checks if session has enough messages (default: 10+)
Pattern Detection: Identifies extractable patterns from the session
Skill Extraction: Saves useful patterns to
```
~/.claude/skills/learned/
```

该技能作为Stop hook在每个会话结束时运行：

会话评估：检查会话是否包含足够数量的消息（默认：10条及以上）
模式检测：从会话中识别可提取的模式
技能提取：将有用的模式保存至
```
~/.claude/skills/learned/
```

Configuration

配置

Edit

config.json

to customize:

json

{
  "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"
  ]
}

编辑

config.json

进行自定义：

json

{
  "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 Types

模式类型

Pattern	Description
`error_resolution`	How specific errors were resolved
`user_corrections`	Patterns from user corrections
`workarounds`	Solutions to framework/library quirks
`debugging_techniques`	Effective debugging approaches
`project_specific`	Project-specific conventions

模式	描述
`error_resolution`	特定错误的解决方式
`user_corrections`	来自用户修正的模式
`workarounds`	框架/库特性问题的解决方案
`debugging_techniques`	有效的调试方法
`project_specific`	项目特定的约定

Hook Setup

Hook设置

Add to your

~/.claude/settings.json

json

{
  "hooks": {
    "Stop": [{
      "matcher": "*",
      "hooks": [{
        "type": "command",
        "command": "~/.claude/skills/continuous-learning/evaluate-session.sh"
      }]
    }]
  }
}

添加至你的

~/.claude/settings.json

：

json

{
  "hooks": {
    "Stop": [{
      "matcher": "*",
      "hooks": [{
        "type": "command",
        "command": "~/.claude/skills/continuous-learning/evaluate-session.sh"
      }]
    }]
  }
}

Why Stop Hook?

为何使用Stop Hook？

Lightweight: Runs once at session end
Non-blocking: Doesn't add latency to every message
Complete context: Has access to full session transcript

轻量：仅在会话结束时运行一次
非阻塞：不会增加每条消息的延迟
完整上下文：可访问完整的会话记录

Comparison Notes (Research: Jan 2025)

对比说明（研究：2025年1月）

vs Homunculus (github.com/humanplane/homunculus)

与Homunculus（github.com/humanplane/homunculus）对比

Homunculus v2 takes a more sophisticated approach:

Feature	Our Approach	Homunculus v2
Observation	Stop hook (end of session)	PreToolUse/PostToolUse hooks (100% reliable)
Analysis	Main context	Background agent (Haiku)
Granularity	Full skills	Atomic "instincts"
Confidence	None	0.3-0.9 weighted
Evolution	Direct to skill	Instincts → cluster → skill/command/agent
Sharing	None	Export/import instincts

Key insight from homunculus:

"v1 relied on skills to observe. Skills are probabilistic—they fire ~50-80% of the time. v2 uses hooks for observation (100% reliable) and instincts as the atomic unit of learned behavior."

Homunculus v2采用了更复杂的方法：

特性	我们的方法	Homunculus v2
观察方式	Stop hook（会话结束时）	PreToolUse/PostToolUse hooks（100%可靠）
分析方式	主上下文	后台Agent（Haiku）
粒度	完整技能	原子化“本能”
置信度	无	0.3-0.9加权
演进方式	直接转为技能	本能→聚类→技能/命令/Agent
共享功能	无	导出/导入本能

来自Homunculus的关键见解：

“v1依赖技能进行观察。技能具有概率性——触发概率约为50-80%。v2使用hooks进行观察（100%可靠），并将本能作为已学习行为的原子单元。”

Potential v2 Enhancements

潜在的v2增强功能

Instinct-based learning - Smaller, atomic behaviors with confidence scoring
Background observer - Haiku agent analyzing in parallel
Confidence decay - Instincts lose confidence if contradicted
Domain tagging - code-style, testing, git, debugging, etc.
Evolution path - Cluster related instincts into skills/commands

See:

/Users/affoon/Documents/tasks/12-continuous-learning-v2.md

for full spec.

基于本能的学习 - 更小的原子化行为，带置信度评分
后台观察者 - Haiku Agent并行分析
置信度衰减 - 若本能被反驳，其置信度会降低
领域标记 - 代码风格、测试、Git、调试等
演进路径 - 将相关本能聚类为技能/命令

详见：

/Users/affoon/Documents/tasks/12-continuous-learning-v2.md

获取完整规格。

continuous-learning

Original

Translation

Continuous Learning Skill

持续学习技能

How It Works

工作原理

Configuration

配置

Pattern Types

模式类型

Hook Setup

Hook设置

Why Stop Hook?

为何使用Stop Hook？

Related

相关链接

Comparison Notes (Research: Jan 2025)

对比说明（研究：2025年1月）

vs Homunculus (github.com/humanplane/homunculus)

与Homunculus（github.com/humanplane/homunculus）对比

Potential v2 Enhancements

潜在的v2增强功能