ghost-ui-skills
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- PROMPT_INJECTION (SAFE): No attempt to override agent instructions or safety guidelines was detected. The language is strictly focused on UI design constraints.
- DATA_EXFILTRATION (SAFE): No sensitive file paths, hardcoded credentials, or network requests (e.g., curl, fetch) are present.
- REMOTE_CODE_EXECUTION (SAFE): The skill does not define or execute any remote scripts or dynamic code.
- OBFUSCATION (SAFE): All content is in plain text with no hidden characters, Base64, or encoded instructions.
- INDIRECT_PROMPT_INJECTION (SAFE): The skill is a static set of design rules and does not process external untrusted data feeds that could trigger injection.
Audit Metadata