← Back to Skills Marketplace
230
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install openclaw-reliable-backup
Description
提供 OpenClaw 自动化每日备份、变更前快照、邮件发送、备份管理及恢复指导，确保数据高可靠性和安全。
Usage Guidance
This skill appears to implement a full backup/restore workflow and will read and archive highly sensitive OpenClaw files (credentials, identity, agents). Before installing or running it: (1) ensure BACKUP_ROOT is explicitly set and has restrictive permissions (chmod 700), (2) prefer encrypting backups (use a passphrase) before storing or emailing them — do not email raw credential-containing archives, (3) review any mail-skill you install manually (clawhub-installed skills could request your email password), (4) test the backup/restore flow in a safe environment first, and (5) ask the skill author to fix metadata inconsistencies (declare BACKUP_ROOT as a required env var) and to add explicit, automated encryption guidance so sensitive data is not accidentally exfiltrated. If you are uncomfortable with backups that include credential files, request a version that omits or redacts those paths or that enforces client-side encryption.
Capability Analysis
Type: OpenClaw Skill
Name: openclaw-reliable-backup
Version: 1.0.0
The skill is designed to archive and potentially exfiltrate highly sensitive data, specifically targeting the `~/.openclaw/credentials/` directory which contains API keys and authentication tokens. It utilizes `command_execution` to create ZIP archives and includes instructions in `SKILL.md` for the agent to dynamically install other skills using `clawhub install`. While these capabilities are aligned with the stated purpose of a comprehensive backup utility, the broad access to secrets and the facilitation of data transmission via email represent a significant attack surface if the agent is influenced by malicious prompts.
Capability Assessment
Purpose & Capability
Skill name/description match the instructions: it explicitly backs up OpenClaw state (many ~/.openclaw paths) and provides restore/lifecycle flows. However the package metadata declares no required environment variables while the SKILL.md treats BACKUP_ROOT as mandatory — a metadata/instruction mismatch that can cause surprises.
Instruction Scope
SKILL.md instructs the agent to read and copy many sensitive paths (e.g. ~/.openclaw/credentials, identity, telegram, agents), create archives, and optionally send them by email. It also directs the agent to install and configure third-party 'mail' skills via commands like `clawhub install` and `openclaw skill config`, and to prompt users for email passwords. Those actions are coherent for a full backup feature but materially expand the attack surface (sensitive secrets bundled into backups; credential entry prompts; installing arbitrary skills).
Install Mechanism
There is no install script or remote download in this skill (instruction-only), which reduces supply-chain risk. The SKILL.md does instruct calling external commands (clawhub/openclaw) to install other skills, but actual code downloads would be performed by those commands, not by this skill itself.
Credentials
Declared requirements list no env vars, but the instructions require BACKUP_ROOT and reference BACKUP_RETENTION_DAYS and EMAIL_RETENTION_DAYS. More importantly, the backup explicitly includes credentials and identity files — appropriate for a full state backup but high sensitivity; backing up and emailing these files without strong encryption or explicit guidance to encrypt is risky.
Persistence & Privilege
Skill is not always:true and is user-invocable; it does not request permanent elevated platform privileges. It does instruct stopping/starting the OpenClaw gateway and installing other skills, which are normal for backup/maintenance but require user consent.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install openclaw-reliable-backup - After installation, invoke the skill by name or use
/openclaw-reliable-backup - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
openclaw-reliable-backup v1.0.0
- Initial release of automated, high-reliability OpenClaw backup and secure management skill
- Features: daily scheduled full backup, pre-modification snapshots, email backups, lifecycle management, and emergency recovery guide
- Interactive setup covers backup directory, email capability detection, and retention strategies
- Supports both local and optional email backup delivery (with guided email skill integration)
- Built-in log tracking, user confirmation on sensitive operations, and detailed error handling
- Emphasizes privacy, user control, and integrity validation for all backups
Metadata
Frequently Asked Questions
What is openclaw-reliable-backup?
提供 OpenClaw 自动化每日备份、变更前快照、邮件发送、备份管理及恢复指导，确保数据高可靠性和安全。 It is an AI Agent Skill for Claude Code / OpenClaw, with 230 downloads so far.
How do I install openclaw-reliable-backup?
Run "/install openclaw-reliable-backup" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is openclaw-reliable-backup free?
Yes, openclaw-reliable-backup is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does openclaw-reliable-backup support?
openclaw-reliable-backup is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created openclaw-reliable-backup?
It is built and maintained by Llmin (@liliwang76); the current version is v1.0.0.
More Skills