overdrive
Pass
Audited by Gen Agent Trust Hub on Mar 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional, focusing on frontend development best practices and the utilization of standard Web APIs (e.g., View Transitions, WebGL, Web Workers).
- [PROMPT_INJECTION]: No patterns indicative of system prompt override, safety filter bypass, or role-play-based jailbreaks were identified.
- [DATA_EXFILTRATION]: The skill does not contain instructions to access sensitive local files (like SSH keys or AWS credentials) or perform unauthorized network requests.
- [REMOTE_CODE_EXECUTION]: There are no commands to download and execute remote scripts or binaries. Mentioned libraries (e.g., Three.js, GSAP) are industry-standard tools suggested for UI enhancement.
- [COMMAND_EXECUTION]: The skill does not invoke dangerous system-level commands or attempt privilege escalation.
- [CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or secrets are present in the skill content.
Audit Metadata