Global Rank · of 601 Skills
byted-tos-openclaw-backup AI Agent Skill
View Source: bytedance/agentkit-samples
MediumInstallation
npx skills add bytedance/agentkit-samples --skill byted-tos-openclaw-backup 12
Installs
OpenClaw 自动备份技能
此技能帮助你自动备份OpenClaw的核心配置、技能源码和记忆数据到挂载的TOS网盘,按日期创建备份目录,自动跳过依赖目录和大体积可执行文件。
何时使用
当用户要求执行以下操作时使用此技能:
- "备份OpenClaw配置"
- "备份技能文件"
- "备份到网盘"
- "创建系统备份"
- "定期备份"
备份策略
备份内容
- 核心配置文件:工作区根目录下所有MD文档(AGENTS.md, SOUL.md, USER.md等)
- 系统配置:
/root/.openclaw/config/目录下的JSON/YAML/ENV/conf等配置文件,跳过二进制文件 - 技能文件:扩展技能和工作区技能的源码文件(JS/TS/PY/MD/JSON/YAML/SH),自动跳过以下目录:
node_modules/venv/__pycache__/dist/build/
- 记忆数据:
/root/.openclaw/workspace/memory/目录下的所有记忆文件
存储结构
my-bucket/
└── openclaw_backup/
├── 2026-03-11/
│ ├── *.md # 核心配置文件
│ ├── config/ # 系统配置
│ ├── skills/
│ │ ├── extensions/ # 扩展技能源码
│ │ └── workspace/ # 工作区技能源码
│ ├── memory/ # 记忆数据
│ ├── backup_summary.txt # 备份统计信息
│ └── backup_manifest.txt # 完整文件清单
└── 2026-03-12/
└── ...指令
1. 执行完整备份
运行备份脚本自动完成所有备份操作:
scripts/backup.sh输出示例:
🔍 检测可用网盘...
✅ 自动选择存储桶: my-bucket (/root/.openclaw/workspace/my-bucket)
ℹ️ 所有可用存储桶:
- my-bucket: /root/.openclaw/workspace/my-bucket
💡 如需固定使用某个存储桶,请运行: bash scripts/config.sh <存储桶名称>
📂 创建备份目录: /root/.openclaw/workspace/my-bucket/openclaw_backup/2026-03-11
📝 备份核心配置文件 (MD文档)...
✅ 核心配置文件备份完成 (7 个文件)
⚙️ 备份系统配置 (仅配置文件,跳过二进制)...
🔍 扫描系统配置子目录...
✅ 备份目录: agents (大小: 228K, 文件数: 5)
✅ 备份目录: canvas (大小: 8.0K, 文件数: 1)
✅ 备份目录: completions (大小: 460K, 文件数: 4)
✅ 备份目录: cron (大小: 12K, 文件数: 2)
⚠️ 跳过目录: extensions (大小: 2.4G, 文件数: 67554 - 超过阈值)
✅ 备份目录: identity (大小: 8.0K, 文件数: 1)
⚠️ 跳过目录: logs (排除列表)
✅ 备份目录: memory (大小: 44K, 文件数: 5)
⚠️ 跳过目录: workspace (排除列表)
✅ 系统配置备份完成 (21 个文件)
🔧 备份技能文件 (仅源码文件,跳过node_modules等依赖目录)...
✅ 技能文件备份完成
🧠 备份记忆数据...
ℹ️ 记忆目录不存在,跳过记忆数据备份
✅ 记忆数据备份完成
📊 生成备份统计...
✅ 备份完成!
📦 备份桶: my-bucket
📂 备份路径: openclaw_backup/2026-03-11
📄 备份文件数: 463
💾 总大小: 3.6M
📋 备份摘要已保存到: backup_summary.txt
📑 完整文件清单已保存到: backup_manifest.txt2. 配置默认存储桶
固定使用某个存储桶进行备份:
scripts/config.sh <存储桶名称>示例:
scripts/config.sh my-bucket3. 检查备份历史
列出所有已创建的备份:
scripts/list_backups.sh输出示例:
📋 现有备份列表 (存储桶: my-bucket):
=====================================
📅 2026-03-11 (463 files, 3.6M)
📅 2026-03-10 (452 files, 3.4M)
📅 2026-03-09 (441 files, 3.2M)4. 恢复备份
从指定日期的备份恢复系统:
scripts/restore.sh <备份日期>预览恢复(不实际修改文件):
scripts/restore.sh 2026-03-11 --dry-run输出示例:
✅ 自动选择存储桶: my-bucket (/root/.openclaw/workspace/my-bucket)
📂 准备恢复备份: 2026-03-11
📦 存储桶: my-bucket
📂 备份路径: /root/.openclaw/workspace/my-bucket/openclaw_backup/2026-03-11
⚠️ 恢复操作会覆盖现有文件,确定要继续吗? (y/N) y
🔄 开始恢复...
📝 恢复核心配置文件...
⚙️ 恢复系统配置...
✅ 系统配置恢复完成
🔧 恢复技能文件...
✅ 技能文件恢复完成
🧠 恢复记忆数据...
✅ 记忆数据恢复完成
✅ 恢复完成!
📅 恢复的备份日期: 2026-03-11
⚠️ 建议重启OpenClaw服务以应用所有配置3. 反馈格式
向用户反馈备份结果时使用以下格式:
"✅ 备份已成功完成!
📦 备份详情
- 存储桶: my-bucket
- 备份路径:
openclaw_backup/2026-03-11/- 备份文件数: 15,076 个
- 总大小: 922 MB
📋 备份内容包含:
- 核心配置文件(工作区所有MD文档)
- 系统配置(JSON/YAML/ENV等配置文件,跳过二进制文件)
- 技能源码(JS/TS/PY/MD等源码文件,自动跳过依赖目录)
- 记忆数据"
注意事项
- 自动检测挂载的网盘,默认使用第一个检测到的存储桶
- 如果未检测到网盘,提示用户在arkClaw界面配置TOS存储桶
- 自动跳过可执行文件、依赖目录和大体积二进制文件
- 每次备份自动创建新的日期目录,不会覆盖历史备份
Installs
Security Audit
View Source
bytedance/agentkit-samples
More from this source
Power your AI Agents with
the best open-source models.
Drop-in OpenAI-compatible API. No data leaves Europe.
Explore Inference APIGLM
GLM 5
$1.00 / $3.20
per M tokens
Kimi
Kimi K2.5
$0.60 / $2.80
per M tokens
MiniMax
MiniMax M2.5
$0.30 / $1.20
per M tokens
Qwen
Qwen3.5 122B
$0.40 / $3.00
per M tokens
How to use this skill
Install byted-tos-openclaw-backup by running npx skills add bytedance/agentkit-samples --skill byted-tos-openclaw-backup in your project directory. Run the install command above in your project directory. The skill file will be downloaded from GitHub and placed in your project.
No configuration needed. Your AI agent (Claude Code, Cursor, Windsurf, etc.) automatically detects installed skills and uses them as context when generating code.
The skill enhances your agent's understanding of byted-tos-openclaw-backup, helping it follow established patterns, avoid common mistakes, and produce production-ready output.
What you get
Skills are plain-text instruction files — not executable code. They encode expert knowledge about frameworks, languages, or tools that your AI agent reads to improve its output. This means zero runtime overhead, no dependency conflicts, and full transparency: you can read and review every instruction before installing.
Compatibility
This skill works with any AI coding agent that supports the skills.sh format, including Claude Code (Anthropic), Cursor, Windsurf, Cline, Aider, and other tools that read project-level context files. Skills are framework-agnostic at the transport level — the content inside determines which language or framework it applies to.
Chat with 100+ AI Models in one App.
Use Claude, ChatGPT, Gemini alongside with EU-Hosted Models like Deepseek, GLM-5, Kimi K2.5 and many more.