frontend-master
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface where user-provided strings are interpolated into system prompts via generated documentation.
- Ingestion points: User-supplied project names and search queries are ingested through CLI arguments in
scripts/design_system.pyandscripts/search.py. - Boundary markers: The skill does not utilize delimiters or explicit instructions to the AI agent to ignore instructions embedded within the interpolated user content when the generated markdown is read back.
- Capability inventory: The skill has the capability to read local design data files and write generated markdown files (
MASTER.mdand page overrides) to the local file system. - Sanitization: While the skill uses normalization and slugification to effectively mitigate path traversal risks in file operations, it does not perform sanitization of the document body content for prompt security before interpolation.
Audit Metadata