OpenClaw(Clawdbot) 是如何“记住一切”的

[0] System Prompt（系统提示词：静态 + 条件指令）
[1] Project Context（项目上下文：启动文件，如 AGENTS.md、SOUL.md 等）
[2] Conversation History（对话历史：消息、工具调用、压缩摘要）
[3] Current Message（当前用户输入）

Context = System Prompt + Conversation History + Tool Results + Attachments

Memory = MEMORY.md + memory/*.md + Session Transcripts

{
  "name": "memory_search",
  "description": "Mandatory recall step: semantically search MEMORY.md + memory/*.md before answering questions about prior work, decisions, dates, people, preferences, or todos",
  "parameters": {
    "query": "What did we decide about the API?",
    "maxResults": 6,
    "minScore": 0.35
  }
}

{
  "results": [
    {
      "path": "memory/2026-01-20.md",
      "startLine": 45,
      "endLine": 52,
      "score": 0.87,
      "snippet": "## API Discussion\nDecided to use REST over GraphQL for simplicity...",
      "source": "memory"
    }
  ],
  "provider": "openai",
  "model": "text-embedding-3-small"
}

{
  "name": "memory_get",
  "description": "Read specific lines from a memory file after memory_search",
  "parameters": {
    "path": "memory/2026-01-20.md",
    "from": 45,
    "lines": 15
  }
}

{
  "path": "memory/2026-01-20.md",
  "text": "## API Discussion\n\nMet with the team to discuss API architecture.\n\n### Decision\nWe chose REST over GraphQL for the following reasons:\n1. Simpler to implement\n2. Better caching\n3. Team familiarity\n\n### Endpoints\n- GET /users\n- POST /auth/login\n- GET /projects/:id"
}

~/clawd/
├── MEMORY.md              # 第二层：长期精选记忆
└── memory/
    ├── 2026-01-26.md      # 第一层：当天日志
    ├── 2026-01-25.md
    ├── 2026-01-24.md
    └── ...

# 2026-01-26

## 10:30 AM - API Discussion
Discussed REST vs GraphQL with user. Decision: use REST for simplicity.
Key endpoints: /users, /auth, /projects.

## 2:15 PM - Deployment
Deployed v2.3.0 to production. No issues.

## 4:00 PM - User Preference
User mentioned they prefer TypeScript over JavaScript.

# Long-term Memory

## User Preferences
- Prefers TypeScript over JavaScript
- Likes concise explanations
- Working on project "Acme Dashboard"

## Important Decisions
- 2026-01-15: Chose PostgreSQL for database
- 2026-01-20: Adopted REST over GraphQL
- 2026-01-26: Using Tailwind CSS for styling

## Key Contacts
- Alice (alice@acme.com) - Design lead
- Bob (bob@acme.com) - Backend engineer

## Every Session

Before doing anything else:
1. Read SOUL.md - this is who you are
2. Read USER.md - this is who you are helping
3. Read memory/YYYY-MM-DD.md (today and yesterday) for recent context
4. If in MAIN SESSION (direct chat with your human), also read MEMORY.md

Don't ask permission, just do it.

1. 文件保存
   ~/clawd/memory/2026-01-26.md
        ↓
2. 文件监听
   Chokidar 监听 MEMORY.md + memory/**/*.md
   1.5 秒防抖
        ↓
3. 分块（Chunking）
   ~400 tokens / chunk，80 tokens 重叠
        ↓
4. 向量化（Embedding）
   每个 chunk → embedding provider → 向量
        ↓
5. 存储
   ~/.clawdbot/memory/<agentId>.sqlite

finalScore = (0.7 * vectorScore) + (0.3 * textScore)

~/.clawdbot/memory/
├── main.sqlite
└── work.sqlite

~/clawd/        # main agent
~/clawd-work/   # work agent

{
  agents: {
    defaults: {
      compaction: {
        reserveTokensFloor: 20000,
        memoryFlush: {
          enabled: true,
          softThresholdTokens: 4000,
          systemPrompt: "Session nearing compaction. Store durable memories now.",
          prompt: "Write lasting notes to memory/YYYY-MM-DD.md; reply NO_REPLY if nothing to store."
        }
      }
    }
  }
}