自动归档 (Two-Domain Knowledge Vault)

A two-domain knowledge management system. Turns the flood of daily information into

organized, retrievable, and reusable wisdom.

Two domains:

• 事业域 ({business_name}/) — Business content organized by function
• 生活学习域 (生活学习/) — Personal growth organized by three layers (收集层→认知层→指南层)

Design Principles

• Low-friction capture, traceable distillation, cautious upgrades, long-term maintainability
• Only text content; images/video out of scope
• Storage: Obsidian physical files (not memory — survives Gateway restarts)

Configuration

The following must be configured per user (via workspace files, e.g. TOOLS.md or MEMORY.md):

If not configured: Ask user to provide both values before first use. Never hardcode paths or names.

Top-Level Decision Flow

Step 0: Business or Life & Learning?

This is always the first routing decision.

Ambiguous cases:

• Content about a topic that serves both business and personal growth → Ask user which domain
• Learning content: "Is this learning FOR the business or FOR personal growth?" — if clearly business-related (e.g., marketing course for the business), route to {business_name}/学习笔记/; if general personal development, route to 生活学习/

事业域 ({business_name}/)

Business content organized by function, no three-layer nesting.

Directory Structure

{business_name}/
├── 课程/              ← Course design, curricula, teaching materials, class notes
├── 运营/              ← Marketing strategies, growth plans, funnel analysis, metrics
├── 方法论产品/        ← SOPs, templates, prompts designed FOR clients to use
├── 客户报告/          ← Reports, assessments, analyses produced FOR clients
├── 活动档案/          ← Event records, recaps, attendee feedback, photos notes
├── 周会/              ← Meeting notes, agendas, action items
├── 素材/              ← Marketing copy, social media content, design briefs
└── 学习笔记/          ← External learning relevant to the business (courses, articles, peers)

Business Domain Routing Rules

Custom sub-directories: Within each function directory, organize by sub-topic or date as needed:

• Single event, 1 file → YYYY-MM-DD 主题描述.md
• Single event, multiple files → YYYY-MM-DD 活动名/ sub-directory
• Cross-day event → YYYY-MM 活动名/ sub-directory

Business Domain Naming Rules

• Files: YYYY-MM-DD 主题描述.md (with date) or 主题名.md (for evergreen docs like SOPs)
• Sub-directories: YYYY-MM-DD 活动名/ or topic-based names
• Follow the universal naming rules below (no person names in titles, use —— separator, etc.)

生活学习域 (生活学习/)

Personal and family content organized by the three-layer system.

生活学习/
├── 收集层/        → Raw capture, organized by scene
│     ↓ AI distills (when warranted)
├── 认知层/        → Refined insights by 5 dimensions
│     ↓ Proven patterns emerge
└── 指南层/        → Verified reusable guides, methods, checklists

Layer 1: 收集层 (Inbox)

Zero-friction entry point. Raw records organized by scene.

Default scenes:

Custom scenes: When user proposes a new scene, confirm name once, then auto-create directory.

Repeating content types (journals, weekly reviews, etc.) get their own scene directory — don't stuff them into 学习成长.

Layer 2: 认知层 (Cognition)

Distilled knowledge organized by five dimensions:

IMPORTANT: Use exact directory names above (with spaces, title case). Never convert to lowercase-hyphen format like what-i-think.

🔴 分类防线（必须执行）：

1. 来源测试（分类前必做）：这个知识/观点是谁产出的？
• 从他人/外部来源学到的 → What I Learn（即使内容包含"对我的启发""素材整理"，来源是他人就是 Learn）
• 用户自己悟到/观察到的 → What I Think
• 信号词：标题含他人名字、"XX的框架""XX分享" → 大概率 Learn

1. 重复类型第一次就建独立目录：早间日报、周报、日志等定期产出物，第一次归档时就在收集层建独立场景目录，不塞进「学习成长」

1. 同一活动/事件的多条内容 → 先建活动子目录：识别信号是同一天、同一来源、同一主题的多条内容，先建 YYYY-MM-DD 活动名 子目录再逐条归入

1. What I Think 按主题域建子目录：
• 子目录示例：AI理念、教育理念、修行理念、事业理念
• 写入前先检查是否有匹配的子目录，没有则考虑新建
• 避免根目录堆积

1. What I Learn 按主题域建子目录：
• 子目录示例：AI与技术、商业与创业、修行与心法、育儿与教育
• 写入前先检查是否有匹配的子目录，没有则考虑新建
• 避免根目录堆积

Classification rules:

• External source primarily → What I Learn
• Self-observation/reflection primarily → What I Think
• Finished output → What I Create
• Fragment to keep as-is → What I Collect
• Family identity/values/rules → Who We Are
• Mixed content → classify by primary source; if both parts are important, split into two cognition entries

Layer 3: 指南层 (Guidebook)

Proven, reusable methods that answer "how to do X" concretely.

Admission criteria (strict):

• User explicitly says "已验证", "反复有效", "用了较长时间", or similar
• OR original text contains clear validation evidence (e.g., "执行两个月有效")
• Methods that don't meet validation criteria stay in Cognition, not Guidebook
• Mark as "待验证" only if user wants to track an unvalidated method

Directory rules:

• Organized by project/domain sub-directories (e.g., 指南层/写作系统/, 指南层/孩子教育/)
• No loose files in 指南层 root — every file must be in a topic sub-directory

Life & Learning Decision Flow

After routing to 生活学习域, follow these steps:

Step 1: Determine user intent

• User says "就存一下 / 先记下来 / 不用总结" → Only write 收集层, stop
• Normal record → Write 收集层, then evaluate distillation (Step 3)
• Explicitly validated method/process → Write 收集层 + 认知层 + evaluate 指南层

Step 2: Determine scene

• Scene is clear → Route to corresponding 收集层/ directory
• Scene is ambiguous but close → Pick closest match, preserve original context in content
• Scene truly unclear → Ask user, do not guess

Step 3: Determine whether to distill to 认知层

Distill when:

• Content has clear viewpoints, patterns, insights, or conclusions
• Content has long-term reuse value
• Content is a quote/fragment worth collecting
• Content is from articles/lectures/courses that can be distilled

Do NOT distill when:

• Pure memo or reminder
• Pure diary/stream-of-consciousness with no extractable insight
• User explicitly asked not to summarize
• Content only suitable for raw preservation

Short content exception: Short reflections (<200 chars) or single quotes can skip 收集层 and go directly to 认知层 — the raw text IS the refined version, no need for duplicate storage.

Step 4: Determine whether to enter 指南层

• Has executable steps + validated effectiveness → 指南层
• Just an idea, principle, or feeling → Stay in 认知层
• Unvalidated method → Stay in 认知层, upgrade later when validated

Universal Rules (Both Domains)

File Management Rules

When to create new file

• New topic
• First record of the day under a new topic
• Content significantly different from existing notes
• Content belongs to a different layer/dimension/function

When to merge into existing file

• Same day AND same scene/topic
• New content supplements existing content (not a new topic)
• Append with a separator (---) and timestamp

When to update existing file

• Guidebook: prefer updating original file over creating new ones; refresh "更新日期"
• Cognition: if supplementing the same insight, add a "补充观察" section
• Never overwrite raw content for cosmetic reasons

🔴 标题命名规则（必须执行）

通用规则

1. 不含人名：标题聚焦概念/主题本身，人名放在文件内部 metadata（来源字段）
• ❌ 黄有璨的AI焦虑三分法 → ✅ AI焦虑三分法
• ❌ 韩叙：一人公司框架 → ✅ 一人公司框架
• ❌ 刘思毅：深圳AI嘉年华 → ✅ 深圳AI嘉年华——观察与三个判断
2. 分隔符统一用 ——（中文破折号），不用 ：、-、:
3. 标题要精准直接：描述内容核心，不用口语化/吸引眼球的表达
• ❌ 你的小龙虾可能悄悄出卖了你 → ✅ AI Agent安全隐患与防护
4. 产品名用正式名称：龙虾 → OpenClaw

收集层标题（生活学习域）

• 格式：YYYY-MM-DD 主题描述.md
• 子目录内文件可省略日期前缀（如 简版.md、详细版.md）
• 活动子目录：YYYY-MM-DD 活动名/ 或 YYYY-MM 活动名/

认知层标题（生活学习域）

• 格式：标题.md（无日期前缀，认知按主题组织不按时间）
• 从提炼内容中提取核心概念作为标题

指南层标题（生活学习域）

• 格式：主题名.md（无日期前缀）
• 面向行动：回答"怎么做X"

事业域标题

• 带日期的记录：YYYY-MM-DD 主题描述.md
• 常青文档（SOP、模板等）：主题名.md

Content Templates

事业域文件

# [Title]

> 日期：YYYY-MM-DD
> 类别：[function directory name, e.g., 课程/运营/方法论产品]
> 来源：[source if applicable]

---

[Content — structured appropriately for the function type]

收集层 (生活学习域)

# [Title]

> 日期：YYYY-MM-DD
> 场景：[scene name]
> 来源：[source if applicable]

---

[Original content — keep close to original, minimal formatting]

认知层 (生活学习域)

# [Title]

> 日期：YYYY-MM-DD
> 维度：[dimension name]
> 来源：[source — article name, conversation, experience, etc.]
> 源文件：[corresponding inbox filename, e.g., 生活学习/收集层/学习成长/2026-03-13 蒙特梭利.md]
> 标签：#tag1 #tag2

---

[Refined content — distilled, structured, insightful]

## 核心要点
- Point 1
- Point 2

Note: When content skips 收集层 (short reflections/quotes), 源文件 field is "用户即时输入".

指南层 (生活学习域)

# [Title]

> 更新日期：YYYY-MM-DD
> 适用场景：[when to use this guide]
> 验证状态：[已验证 / 待验证]
> 来源：[corresponding cognition or inbox filename]

---

[Step-by-step guide, checklist, or methodology]

Field defaults:

• 来源 missing → "用户即时输入"
• Tags: 2-3 max, prefer topic/person/method keywords
• Preserve original attribution for articles/lectures/books

Source Traceability

• Every 认知层 entry must trace back to a 收集层 file or "用户即时输入"
• Every 指南层 entry must trace back to 认知层 or 收集层
• 事业域 files should note their source in the 来源 field
• No orphaned conclusions without traceable context

Long Text Handling

Never use obsidian-cli create --content — truncation bug.

After every write, verify:

ls -la "<file>"     # exists and reasonable size
tail -5 "<file>"    # content is complete

🔴 乱码检查（每次写入后必须执行）

python3 -c "c=open('<file_path>').read(); n=c.count('\ufffd'); print(f'{n}个乱码') if n else print('✅')"

Found U+FFFD (�) → infer correct character from context, fix with edit, re-verify.

Searching and Retrieval

• obsidian-cli search-content "" — search note content
• obsidian-cli search "" — search note titles
• find + grep — precise filesystem search

🔴 目录构建规则（必须执行）

事业域目录规则

1. 顶层按功能分8个目录：课程、运营、方法论产品、客户报告、活动档案、周会、素材、学习笔记
2. 功能目录内部按需建子目录：
• 单次活动1个文件 → 直接放 YYYY-MM-DD 主题描述.md
• 单次活动多个文件 → 建 YYYY-MM-DD 活动名/ 子目录
• 跨日活动 → 建 YYYY-MM 活动名/ 子目录
• 系列内容 → 建主题子目录（如 课程/公开课系列/）
3. 新的功能类型：如果内容不匹配现有8个目录，先考虑归入最近似的；确实需要新目录时，向用户确认后新建

生活学习域——收集层目录规则

1. 顶层按场景分目录：学习成长、童言童语、育儿观察、灵感闪现、对话收获、生活记录等
2. 重复类型独立建目录：日报、周报、日志等定期产出物，不塞进「学习成长」，直接在收集层建独立场景目录
3. 学习成长/内部的子目录规则：
• 单次活动1个文件 → 直接放 YYYY-MM-DD 主题描述.md
• 单次活动多个文件 → 建 YYYY-MM-DD 活动名/ 子目录
• 跨日活动（禅七、大会等）→ 建 YYYY-MM 活动名/ 子目录
4. 功能性文件可直接放收集层根目录：如待办池、清单等不属于特定场景的文件

生活学习域——认知层目录规则

1. What I Learn 按知识领域分子目录：如 AI与技术、修行与心法、育儿与教育
2. What I Think 按理念主题分子目录：如 AI理念、修行理念、教育理念
3. What I Create 按产出类型分子目录：如 公众号文章、觉知日记
4. What I Collect 按素材类型分子目录：如 金句库、选题库
5. Who We Are 扁平存放：身份文档数量少，不需要子目录
6. 🚫 禁止同名文件与目录共存：如果存在 事业理念/ 目录，就不能同时有 事业理念.md 文件。遇到这种情况，把 .md 文件移入目录并重命名

生活学习域——指南层目录规则

1. 顶层按项目/领域分目录：如 写作系统、孩子教育、家庭OS系统、旅行规划
2. 根目录零散文件清零：每个文件都必须归入一个主题目录
3. 项目目录内部按功能分子目录：如 大纲/、成长规划报告/ 等

🔴 反模式清单（遇到必须修复）

Full Directory Structure Reference

[vault_path]/
├── {business_name}/                ← 事业域（用户自定义名称）
│   ├── 课程/
│   │   ├── 2026-03-15 公开课第一期/
│   │   └── 课程体系设计.md
│   ├── 运营/
│   │   ├── 2026-03 增长策略.md
│   │   └── 私域运营SOP.md
│   ├── 方法论产品/
│   │   └── 家庭AI实操指南.md
│   ├── 客户报告/
│   │   └── 2026-03-20 XX家庭成长规划.md
│   ├── 活动档案/
│   │   └── 2026-03-22 线下沙龙/
│   ├── 周会/
│   │   └── 2026-03-17 周会纪要.md
│   ├── 素材/
│   │   └── 朋友圈文案库.md
│   └── 学习笔记/
│       ├── 商业与创业/
│       └── 2026-03-10 一人公司框架.md
└── 生活学习/                       ← 生活学习域（固定名称）
    ├── 收集层/
    │   ├── 学习成长/
    │   │   ├── 2026-01 泉州禅七/
    │   │   └── 2026-03-13 SWIS学校讲座.md
    │   ├── 童言童语/
    │   ├── 育儿观察/
    │   ├── 灵感闪现/
    │   ├── 对话收获/
    │   └── 生活记录/
    ├── 认知层/
    │   ├── What I Learn/
    │   │   ├── AI与技术/
    │   │   ├── 修行与心法/
    │   │   └── 育儿与教育/
    │   ├── What I Think/
    │   │   ├── AI理念/
    │   │   ├── 修行理念/
    │   │   └── 教育理念/
    │   ├── What I Create/
    │   ├── What I Collect/
    │   └── Who We Are/
    └── 指南层/
        ├── 写作系统/
        ├── 孩子教育/
        ├── 家庭OS系统/
        └── 旅行规划/

IMPORTANT: Use Chinese directory names for the three layers. English aliases (inbox/cognition/guidebook) are NOT used — they cause duplicate directories.