10 KiB

Raw Permalink Blame History

Core Memory Index

Identity & Role

Name: Eason (陈医生)
Role: Professional AI Agent Architect and Administrator
Primary Goals:
1. Improve operational efficiency (speed, cost, resource usage)
2. Enhance work accuracy
3. Strengthen system security, stability, scalability, and portability
4. Maintain comprehensive documentation of all configurations, troubleshooting, debugging, optimization, and new tools

File Structure Overview

/root/.openclaw/workspace/
├── CORE_INDEX.md          # This file - memory index (Layer 1)
├── IDENTITY.md           # Agent identity definition (Layer 1)
├── SOUL.md              # Core personality and behavior guidelines (Layer 1)
├── USER.md              # Information about the human user (Layer 1)
├── MEMORY.md             # Long-term curated memories and decisions (Layer 2)
├── AGENTS.md            # Agent operations and logging practices
├── TOOLS.md             # Environment-specific tool configurations
├── HEARTBEAT.md         # Periodic check tasks
├── agents.yaml          # Central agent registry (config-driven: deploy.sh, agent-monitor.js)
├── deploy.sh            # One-click deployment & management script (config-driven, reads agents.yaml)
├── agent-monitor.js     # Auto-healing & health monitoring system (config-driven, reads agents.yaml)
├── memory/              # Daily memory files YYYY-MM-DD.md (Layer 2)
├── docs/                # Architecture & reference documentation
│   ├── CONTROL_UI_ACCESS_AND_SECURITY.md  # ★ Control UI 访问与安全 (Tailscale+HTTPS+Token+Approve)
│   ├── EXTENSIONS_ARCHITECTURE.md  # ★ 自定义扩展权威参考 (监控+记忆+部署)
│   ├── MULTI_AGENT_MANAGEMENT.md   # ★ 多 Agent 管理 (Hub-Spoke, Onboarding, 远程)
│   ├── MEMORY_ARCHITECTURE.md      # 四层记忆体系详细文档 (v2.1)
│   ├── MEM0_ARCHITECTURE.md        # (旧版, 已废弃 → 见 MEMORY_ARCHITECTURE.md)
│   ├── SYSTEM_ARCHITECTURE.md      # 系统总体架构
│   └── ...
├── skills/              # Installed agent skills
│   └── mem0-integration/        # Layer 4 记忆系统核心 Skill
│       ├── SKILL.md             # Skill 开发者文档 (含 API 规范)
│       ├── mem0_client.py       # 核心客户端 (检索 / 写入 / 队列)
│       ├── openclaw_interceptor.py  # Pre/Post-Hook 拦截器
│       ├── local_search.py      # Layer 3 FTS5 本地检索 fallback
│       ├── memory_cleanup.py   # Memory cleanup & audit (--dry-run / --execute)
│       ├── config.yaml          # mem0 配置
│       └── project_registry.yaml    # Agent-项目归属 (可见性控制)
├── scripts/             # Scripts and utilities
│   ├── parse_agents.py  # Agent registry parsing helper (reads agents.yaml)
│   ├── setup-cron.sh    # Install/remove automated backup + cleanup cron jobs
│   └── 10-create-backup.sh  # Standalone backup script (secondary)
├── templates/           # Agent onboarding templates
│   ├── onboard.sh               # New agent creation script
│   ├── offboard.sh              # Agent offboarding and cleanup script
│   ├── agent-workspace/          # Workspace file templates
│   └── systemd/                  # Service & env file templates
├── logs/                # Operation and system logs
│   ├── operations/      # Manual operations and changes
│   ├── system/          # System-generated logs
│   ├── agents/          # Individual agent logs  
│   └── security/        # Security operations and audits
└── systemd/             # Systemd service definitions & env files
    ├── openclaw-gateway-user.service      # 用户级 Gateway 模板
    ├── openclaw-agent-monitor.service     # 系统级 Monitor 模板
    ├── openclaw-gateway.service.legacy    # 废弃的系统级 Gateway (已 masked)
    └── gateway.env                        # Gateway 环境变量 (升级安全)

Memory Access Strategy

Core Index: Always loaded first - provides structural overview
Lazy Loading: Load specific documents only when needed
Context Injection: Relevant documents passed as context for specific tasks
Version Control: All critical files tracked in Git with rollback capability

Key Documentation Files

★ Control UI 访问与安全: docs/CONTROL_UI_ACCESS_AND_SECURITY.md → Tailscale 内网 + HTTPS + Token + 首次设备 Approve 标准流程；新增 Agent / 迁移时的 UI 与安全配置
★ Extensions Architecture: docs/EXTENSIONS_ARCHITECTURE.md
★ Multi-Agent Management: docs/MULTI_AGENT_MANAGEMENT.md → Part A: 架构参考 (Hub-Spoke, 远程 Agent); Part B: 操作手册 (Sec 11-16) — 创建/维护/记忆管理/移除 Agent 的交互式 Playbook + 备份恢复 (Sec 15) + 服务器迁移 (Sec 16)
Memory Architecture: docs/MEMORY_ARCHITECTURE.md → 四层记忆体系详细设计 (v2.1)
Skill Developer Guide: skills/mem0-integration/SKILL.md → Layer 4 代码结构、API 规范、开发者注意事项
Security Templates: MEMORY.md → Server security hardening templates
Agent Practices: AGENTS.md → Agent deployment and management practices
Logging Standards: AGENTS.md → Operation logging and audit practices
Health Monitoring: agent-monitor.js → Auto-healing, crash detection, Telegram notifications
Deployment: deploy.sh → One-click install/start/stop/rollback/backup/restore/debug/fix-service (config-driven, reads agents.yaml)
Systemd Services: systemd/*.service + *.env → 服务定义及升级安全环境变量
Configuration Backup: Git commits before any JSON modifications

Usage Instructions for Models

Read CORE_INDEX.md first to understand available resources
Identify relevant documentation based on task requirements
Load specific files using read/edit/write tools as needed
Never assume memory persistence across model sessions
Always verify current state before making changes
修改基础设施前 (systemd、监控、部署脚本、环境变量)，必须先读 docs/EXTENSIONS_ARCHITECTURE.md
OpenClaw 升级后，运行 ./deploy.sh fix-service && ./deploy.sh restart 恢复自定义配置
创建新 Agent 前，必须先读 docs/MULTI_AGENT_MANAGEMENT.md Section 11 (Onboarding Playbook)，按对话流程逐步收集信息后执行
Control UI 访问 / 新设备 Approve / 迁移 → 读 docs/CONTROL_UI_ACCESS_AND_SECURITY.md，按 Tailscale+HTTPS+Token+Approve 标准配置
维护/排查 Agent → Section 12; 记忆管理 → Section 13; 移除 Agent → Section 14
备份/恢复 → Section 15; 服务器迁移 → Section 16
定期备份: 运行 scripts/setup-cron.sh 安装自动定时备份 (每天 02:00) + 记忆清理 (每周日 03:00)

System Architecture (2026-02-20)

Layer 1: System-Level (Systemd)

openclaw-gateway.service: Main OpenClaw gateway with auto-restart
openclaw-agent-monitor.service: Health monitoring & auto-healing
Features: Boot auto-start, crash recovery, resource limits, watchdog

Layer 2: Memory Architecture

Core Memory: CORE_INDEX.md - Always loaded first (identity, structure, index)
Long-term Memory: MEMORY.md - Curated decisions, security templates, configs
Daily Memory: memory/YYYY-MM-DD.md - Raw conversation logs, auto-saved
Passive Archive: Convert valuable conversations to skills/notes on request

Layer 3: Version Control (Git)

Repository: /root/.openclaw/workspace
Features: One-click rollback, backup before changes, commit history
Commands: ./deploy.sh rollback, ./deploy.sh backup, ./deploy.sh rollback-to <commit>

Layer 4: Monitoring & Notifications

Health Checks: Every 30 seconds (gateway status, memory, disk)
Auto-Healing: Automatic restart on crash (max 5 restarts per 5 min)
Notifications: Telegram alerts on critical events (stop/error/restart)
Logging: Comprehensive logs in /logs/agents/health-YYYY-MM-DD.log

Management Commands

./deploy.sh install        # Install & start all services
./deploy.sh status         # Check service status (gateway + monitor)
./deploy.sh health         # Run health check
./deploy.sh logs           # View recent logs
./deploy.sh backup         # Full backup (workspace + Qdrant snapshot + profiles)
./deploy.sh backup quick   # Quick backup (workspace only)
./deploy.sh restore <dir>  # Restore workspace from backup
./deploy.sh restore-qdrant <file>  # Restore Qdrant from snapshot
./deploy.sh rollback       # Rollback to previous commit
./deploy.sh debug-stop     # Stop ALL services (safe for debugging)
./deploy.sh debug-start    # Restore all services after debugging
./deploy.sh fix-service    # Re-inject EnvironmentFile after UI upgrade

Memory Architecture (四层记忆体系)

详细文档: docs/MEMORY_ARCHITECTURE.md

Memory Layer 1: Core Memory (核心记忆)

CORE_INDEX.md, IDENTITY.md, SOUL.md 等 MD 文件
启动时首先加载，定义 Agent 身份和行为准则
每个 Agent 独立工作区

Memory Layer 2: Daily Memory (日常记忆)

MEMORY.md (长期策略) + memory/*.md (每日记录)
Git 版本控制保护，支持回溯

Memory Layer 3: Short-term Memory (短期记忆 / QMD)

SQLite FTS5 全文检索 (零额外内存，纯离线)
可选 GGUF 本地向量 (按需加载，需 >= 300MB 空闲内存)
Layer 4 不可达时自动接管检索

Memory Layer 4: Mem0 Conversation Memory (对话记忆)

Qdrant (mem0_v4_shared) + text-embedding-v4 (1024 维)
通过 Tailscale 可跨服务器共享
三级可见性: public (集群共享) / project (项目共享) / private (仅自身)
记忆衰减: session=7天, chat_summary=30天, preference/knowledge=永久
智能写入过滤: 自动跳过简短确认、系统命令等无价值对话

Memory Visibility (记忆可见性)

级别	写入方式	检索范围
public	`visibility=public`	所有 Agent
project	`visibility=project, project_id=<项目>`	同项目 Agent
private	`visibility=private`	仅自身

项目注册表: skills/mem0-integration/project_registry.yaml

10 KiB Raw Permalink Blame History

Core Memory Index

Identity & Role

File Structure Overview

Memory Access Strategy

Key Documentation Files

Usage Instructions for Models

System Architecture (2026-02-20)

Layer 1: System-Level (Systemd)

Layer 2: Memory Architecture

Layer 3: Version Control (Git)

Layer 4: Monitoring & Notifications

Management Commands

Memory Architecture (四层记忆体系)

Memory Layer 1: Core Memory (核心记忆)

Memory Layer 2: Daily Memory (日常记忆)

Memory Layer 3: Short-term Memory (短期记忆 / QMD)

Memory Layer 4: Mem0 Conversation Memory (对话记忆)

Memory Visibility (记忆可见性)

10 KiB

Raw Permalink Blame History