The Agent Skills Directory

[SAFE]: The skill's primary function is as a runtime security monitor for the OpenClaw environment. It provides a comprehensive set of rules and patterns to detect common attack vectors in other skills, including unauthorized credential access, persistence mechanisms, and suspicious network activity.
[SAFE]: The permission configuration is highly restrictive and appropriate for its stated purpose. It declares 'fileRead' as true while explicitly disabling 'fileWrite', 'network', and 'shell' access, ensuring it cannot perform the actions it is designed to monitor.
[SAFE]: No evidence of malicious behavior was found. All references to sensitive files (e.g., SSH keys, environment variables) and dangerous commands (e.g., curl, eval) are documented as suspicious patterns to be flagged in other skills, rather than being part of this skill's own executable instructions.

skill-guard