refactoring-ui
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and review user-provided CSS, SCSS, and HTML code, which acts as an ingestion point for untrusted data. Malicious instructions could be embedded within code comments or text segments in the files being audited, potentially influencing the agent's behavior.
- Ingestion points: User-supplied CSS, SCSS, and HTML snippets or files processed during the 'Design Review' mode or by the
scripts/audit_css.pyscript. - Boundary markers: The instructions do not define specific delimiters or warnings to isolate the processed content from the agent's instruction context, nor do they specify that embedded text should be ignored as instructions.
- Capability inventory: The agent has the capability to read files and execute the provided Python audit script, which accesses and outputs information based on the content of the analyzed files.
- Sanitization: No sanitization or filtering mechanism is described to neutralize potential injection patterns within the code being analyzed.
Audit Metadata