skill-vetter

To assess OpenClaw skill safety, run a vetting process that performs metadata verification and permission risk assessment. It assigns a trust score and provides a clear recommendation on whether to install, review further, or block the skill.

A permission risk assessment for OpenClaw skills analyzes the requested permission scopes to identify unauthorized access attempts. It checks for suspicious patterns and critical warnings to ensure the skill does not pose a security threat to your system.

To check an unfamiliar GitHub skill for typosquatting, run a security-first vetting process that verifies metadata and detects typosquats. This identifies suspicious patterns and red flags to mitigate potential risks from unfamiliar sources.

No special environment setup is required. The vetting process operates directly on the skill file, analyzing metadata, permission scopes, and code content to identify malicious code, unauthorized access attempts, and suspicious patterns.

Yes, you can get a trust score for a new OpenClaw skill before installation. The vetting process assigns a trust score based on metadata verification, permission risk assessment, and content analysis, yielding a clear install or block recommendation.