ClawDog Backup for Openclaw

A robust automated backup and restoration utility for OpenClaw core configuration files and memory layers via OneDrive.

michael-crazy
v1.0.1
Feb 19, 2026
0
1.5k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install clawdog-backup

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install clawdog-backup using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is ClawDog Backup?

ClawDog Backup is a specialized synchronization tool designed to ensure the persistence and safety of your AI agent's identity and knowledge base. By integrating with OneDrive, this skill provides a reliable safety net for the core components of Openclaw Skills, protecting vital files like SOUL.md, AGENTS.md, and the entire memory directory from data loss.

This utility leverages industry-standard tools like rclone and inotify-tools to create a hybrid backup environment. It offers both real-time monitoring for immediate configuration changes and scheduled snapshots for long-term memory retention, making it an essential tool for developers who require high availability and easy migration for their agentic workflows.

ClawDog Backup Use Cases

  • Real-time protection of core agent identity and configuration files against accidental deletion.
  • Automated weekly snapshots of extensive memory directories to maintain a historical record of agent learning.
  • Simplified migration of OpenClaw environments across different machines or server instances.
  • Disaster recovery using dry-run verification before performing full state restorations.

How ClawDog Backup Works

  1. Monitoring: The skill utilizes inotify-tools to monitor core files (SOUL.md, USER.md, etc.) for any modifications in real-time.
  2. Archiving: Upon detecting a change or reaching a scheduled time, the skill packages the relevant files into a compressed tar.gz archive.
  3. Cloud Synchronization: It uses rclone to securely upload the archives to a designated OneDrive path at /Backup/OpenClaw/.
  4. Logging: Detailed logs are generated for every backup and restore operation to ensure auditability.
  5. Restoration: A dedicated recovery script allows users to pull specific versions of their agent's state from the cloud or local storage back into the workspace.

ClawDog Backup Setup

To utilize this skill, ensure rclone is configured with your OneDrive account and inotify-tools is installed on your system.

# Start the real-time core file monitor in the background
nohup /root/.openclaw/workspace/skills/clawdog-backup/scripts/backup-core.sh > /root/.openclaw/backup/inotify.log 2>&1 &

# Manually trigger a memory layer backup
/root/.openclaw/workspace/skills/clawdog-backup/scripts/backup-memory.sh

ClawDog Backup Data Schema & Taxonomy

The skill organizes data into compressed archives with standardized naming conventions for easy identification:

Data Type Target Files Schedule/Trigger
Core Layer SOUL.md, AGENTS.md, IDENTITY.md, USER.md, TOOLS.md Real-time (inotify)
Memory Layer memory/ directory Weekly (Sunday 23:00 UTC)

Local logs are maintained in /root/.openclaw/backup/ including core-backup.log and memory-backup.log.

ClawDog Backup Advanced Features

  • Real-time inotify monitoring for immediate synchronization of configuration changes.
  • Flexible restoration options including --core, --memory, or --all flags.
  • Support for local source restoration, allowing recovery even without active rclone configurations.
  • Dry-run capability (--dry-run) to simulate restoration and prevent accidental overwriting of existing workspace data.
  • Automated cron integration for hands-off weekly memory maintenance within the Openclaw Skills ecosystem.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*