OpenClaw Daily Backup

WarnAudited by ClawScan on May 18, 2026.

Overview

The backup purpose is coherent, but the package includes pre-existing OpenClaw backup snapshots and gives conflicting safety guidance about secret-containing backups.

Install only if you are comfortable with a tool that reads and overwrites core OpenClaw identity/config files. Before restoring, inspect or remove the bundled backups directory, create your own backup, use --dry-run, and do not commit or share backups unless you have confirmed secrets are removed or the backups are encrypted.

Findings (4)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

If you run restore before creating and selecting your own backup, your agent's identity, instructions, user profile, or tool configuration could be replaced with bundled backup content.

Why it was flagged

The release includes pre-existing backup snapshots of core agent instruction, identity, user-profile, and configuration files. Because the skill documents restoring the latest backup, these bundled files could become persistent workspace context if restored.

Skill content
backups/2026-03-14T05-39-50/AGENTS.md ... backups/2026-03-14T05-39-50/SOUL.md ... backups/2026-03-14T05-39-50/USER.md ... backups/2026-03-14T05-39-50/openclaw.sanitized.json
Recommendation

Do not restore until you have inspected the backups directory. The publisher should remove bundled backups from the release, ignore sample backups by default, and require an explicit user-created backup selection.

What this means

You could accidentally expose API keys, tokens, or private user information if you publish or sync the backup directory assuming it has been fully sanitized.

Why it was flagged

This broad safety claim conflicts with the documented behavior that TOOLS.md is backed up and SKILL.md states backups may contain API keys and user information. The mismatch can cause users to over-trust backups as safe to share or commit.

Skill content
No API keys or tokens are stored in backup files
Recommendation

Treat all backups as sensitive unless independently verified. Encrypt them before off-machine storage and avoid committing them to public or shared repositories.

What this means

A mistaken restore can roll your agent configuration back to an older or unintended state.

Why it was flagged

The restore command is intended to overwrite core workspace files. This is central to the skill's backup/recovery purpose and dry-run guidance is documented, but it remains a high-impact local mutation.

Skill content
# Restore latest backup
node scripts/restore.mjs
Recommendation

Always run the documented --dry-run command first, inspect the target backup, and keep the automatic pre-restore backup for rollback.

What this means

Backups may continue running on a schedule after setup and may keep creating sensitive copies of workspace files.

Why it was flagged

The skill documents recurring scheduled execution through cron or heartbeat. This is disclosed and appropriate for daily backups, but it creates persistent background behavior if the user installs it.

Skill content
Add to OpenClaw heartbeat or system cron: ... 0 2 * * * cd /Users/m1/.openclaw/workspace-hunter/soul-backup-skill && node scripts/backup.mjs
Recommendation

Only add the cron/heartbeat entry if you want ongoing backups, and periodically review scheduled jobs and backup retention.