← Back to Skills Marketplace
Dont Hack Me
by
peterokase42
· GitHub ↗
· v1.0.1
5630
Downloads
31
Stars
31
Active Installs
2
Versions
Install in OpenClaw
/install dont-hack-me
Description
別駭我！基本安全檢測 — Security self-check for Clawdbot/Moltbot. Run a quick audit of your clawdbot.json to catch dangerous misconfigurations — exposed gateway, missing auth, open DM policy, weak tokens, loose file permissions. Auto-fix included. Invoke: "run a security check" or "幫我做安全檢查".
Usage Guidance
Install only if you want an agent to inspect your Clawdbot configuration. Invoke it with a specific request like "run dont-hack-me against my Clawdbot config," avoid sharing the full report if it includes secrets, and review each proposed fix before approving changes to auth, tokens, gateway bind, DM/group policy, or file permissions.
Capability Analysis
Type: OpenClaw Skill
Name: dont-hack-me
Version: 1.0.1
The skill performs a security self-check and auto-fix for the agent's own configuration file (`~/.clawdbot/clawdbot.json`). All actions, including reading the config, checking file permissions with `stat`, generating tokens with `openssl`, and modifying the config with `chmod`, are directly aligned with the stated purpose of auditing and hardening the agent's security. The skill explicitly instructs the agent to ask for user consent before applying any fixes and to back up the configuration file (`cp ~/.clawdbot/clawdbot.json ~/.clawdbot/clawdbot.json.bak`) before making changes. There is no evidence of data exfiltration, malicious execution, persistence, or prompt injection attempting to subvert the agent for harmful purposes.
Capability Assessment
Purpose & Capability
The stated purpose is to audit and harden Clawdbot/Moltbot security settings, and the file reads, permission checks, token generation, chmod, backup, and config edits are aligned with that purpose.
Instruction Scope
The runtime checks are scoped to ~/.clawdbot/clawdbot.json, but invocation phrases such as "run a security check", "check my security settings", and "am I secure?" are broad enough to trigger before the user has clearly specified this skill or this config file.
Install Mechanism
The package contains only SKILL.md, with no executable scripts, helper binaries, dependency install flow, or hidden install-time behavior.
Credentials
Reading a local agent config that may contain gateway tokens, passwords, API keys, or private keys is proportionate for this audit, but it is sensitive and should happen only in an explicitly trusted session.
Persistence & Privilege
There is no background persistence or external exfiltration behavior, but approved fixes persistently modify gateway binding, authentication, token, channel policy, and file permissions after a backup.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install dont-hack-me - After installation, invoke the skill by name or use
/dont-hack-me - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
加入中文 description「別駭我！基本安全檢測」、author 補回小安台灣、footer 補回中文標語
v1.0.0
Initial release — adds a security self-check skill for Clawdbot/Moltbot.
- Audits 7 critical areas of `clawdbot.json` (gateway bind, auth, token strength, DM policy, group policy, file permissions, plaintext secrets).
- Reports issues as PASS, FAIL, WARN, or SKIP — with summary scoring.
- Offers auto-fix flow for most misconfigurations, including safe token and permissions fixes.
- Generates a clear, formatted security report for user review.
- Guides users through safe remediation steps, including pre-change backups.
Metadata
Frequently Asked Questions
What is Dont Hack Me?
別駭我！基本安全檢測 — Security self-check for Clawdbot/Moltbot. Run a quick audit of your clawdbot.json to catch dangerous misconfigurations — exposed gateway, missing auth, open DM policy, weak tokens, loose file permissions. Auto-fix included. Invoke: "run a security check" or "幫我做安全檢查". It is an AI Agent Skill for Claude Code / OpenClaw, with 5630 downloads so far.
How do I install Dont Hack Me?
Run "/install dont-hack-me" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Dont Hack Me free?
Yes, Dont Hack Me is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Dont Hack Me support?
Dont Hack Me is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Dont Hack Me?
It is built and maintained by peterokase42 (@peterokase42); the current version is v1.0.1.
More Skills