focus-styles
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon external stylesheets and component code provided by the user, which could potentially contain adversarial instructions disguised as comments or data.
- Ingestion points: CSS files, component styles, and rendered layout definitions are processed in the 'Check' and 'Code Review' sections of SKILL.md.
- Boundary markers: The instructions lack explicit delimiters or guidance for the agent to treat ingested code as untrusted data or to ignore embedded instructions.
- Capability inventory: The agent has the authority to propose code modifications ('Fix'), evaluate selectors ('Code Review'), and generate explanations based on the input.
- Sanitization: No mechanisms for filtering, escaping, or validating the external CSS content are specified.
Audit Metadata