personal-trainer-master
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides industry-grade instructions for exercise programming and client management. The included bash scripts (e.g.,
agentic.sh,new-client-onboarding.sh) are local interactive tools that generate Markdown reports based on user input. No remote network calls or unauthorized file access patterns were found.\n- [SAFE]: The 'Agentic Protocol' correctly instructs the agent to use official search and fetch tools for fact-checking when necessary, which is standard behavior for AI agents and does not constitute a security risk.\n- [SAFE]: No obfuscation, base64-encoded payloads, or suspicious unicode characters were found in the skill metadata, instructions, or scripts.\n- [SAFE]: The skill strictly defines its 'Honest Boundaries' and 'Scope of Practice', explicitly prohibiting the agent from diagnosing medical conditions or prescribing clinical diets, which aligns with safety best practices for fitness-related AI skills.
Audit Metadata