What should I do if my gateway is compromised?

Immediately stop the gateway, rotate all authentication tokens and provider API keys, and run a deep security audit to collect logs and identify the breach source.

How does the security audit work?

The audit command checks inbound access policies, tool blast radius, network exposure, file permissions, and identifies any 'dangerous' configuration flags that bypass safety checks.

How do I manage API keys securely in OpenClaw?

Use the SecretRef system to reference credentials from environment variables, files, or external commands (like a vault) rather than hardcoding them in configuration files.

Can I use OpenClaw over the public internet?

While possible, it is highly discouraged. It is recommended to use 'loopback' or 'tailscale' modes. Using 'funnel' exposes the gateway to the public internet and triggers a critical audit finding.

What is the OpenClaw trust model?

OpenClaw is designed as a personal assistant with a single trusted operator boundary per gateway. For mixed-trust environments, it is recommended to split boundaries using separate gateways or OS users.

OpenClaw Security

Name: OpenClaw Security
Author: TheophilusChinomona

byTheophilusChinomona

0•

보안 및 테스팅

Secures and audits OpenClaw self-hosted AI gateways through hardening, secrets management, and automated incident response.

OpenClaw Security is a specialized Claude Code skill designed to protect and maintain the integrity of self-hosted AI gateway environments. It provides a comprehensive toolkit for performing deep security audits, configuring network isolation via Tailscale or reverse proxies, and managing sensitive API credentials through secure references (SecretRef). Whether you are hardening a new installation, managing trust boundaries, or responding to a potential security incident, this skill provides the domain-specific guidance and commands needed to maintain a robust security posture for your AI infrastructure.

주요 기능

01Comprehensive secrets management using environment variables, files, or external vaults

020 GitHub stars

03Automated file permission checks and remediation for configuration data

04Deep security auditing with automated fix suggestions for common vulnerabilities

05Step-by-step incident response workflows to contain and remediate breaches

06Network hardening configurations for loopback, LAN, and Tailscale VPN modes

사용 사례

01Hardening a self-hosted OpenClaw gateway for remote access via Tailscale

02Performing periodic security audits to detect configuration drift or insecure flags

03Rotating leaked API keys and regenerating gateway authentication tokens during an incident

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add theophiluschinomona/claude-openclaw-plugin openclaw-security

For use in Claude.ai and ChatGPT

주요 기능

01Comprehensive secrets management using environment variables, files, or external vaults

020 GitHub stars

03Automated file permission checks and remediation for configuration data

04Deep security auditing with automated fix suggestions for common vulnerabilities

05Step-by-step incident response workflows to contain and remediate breaches

06Network hardening configurations for loopback, LAN, and Tailscale VPN modes

사용 사례

01Hardening a self-hosted OpenClaw gateway for remote access via Tailscale

02Performing periodic security audits to detect configuration drift or insecure flags

03Rotating leaked API keys and regenerating gateway authentication tokens during an incident

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add theophiluschinomona/claude-openclaw-plugin openclaw-security

For use in Claude.ai and ChatGPT