frontend-design
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security vulnerabilities, malicious patterns, or data exfiltration risks were identified in the analyzed files.
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill does not perform any network operations, access local sensitive files, or handle credentials.
- [REMOTE_CODE_EXECUTION]: No remote script downloads or dynamic execution patterns (like eval or subprocess calls) are present.
- [PROMPT_INJECTION]: Instructional markers like 'CRITICAL' and 'IMPORTANT' are used strictly for design quality guidance and do not attempt to override system safety protocols or extract system prompts.
- [INDIRECT_PROMPT_INJECTION]: While the skill interprets user-provided requirements to generate code, it does not possess capabilities that could be exploited via injected instructions in that data, nor does it interpolate data into unsafe execution contexts within its own logic.
Audit Metadata