normalize
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is focused on design consistency and code refactoring. It instructs the agent to search for local design system documentation and apply consistent styles, tokens, and components to the codebase.
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests and processes untrusted data from the local repository.
- Ingestion points: Reads local design system documentation, UI guidelines, and feature source code (SKILL.md).
- Boundary markers: None identified. The skill does not provide specific delimiters to separate data from instructions during analysis.
- Capability inventory: The agent is authorized to read files (via grep) and modify or delete files as part of the normalization and cleanup process (SKILL.md).
- Sanitization: None identified. The skill relies on the agent's internal reasoning to differentiate between code structure and embedded instructions.
- [SAFE]: The skill references a legitimate external source from a well-known developer (Paul Bakaus) for its logic, which is documented neutrally in the metadata.
Audit Metadata