shape
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests project-level files to inform design planning. \n
- Ingestion points: The skill reads existing context from
.design-context.mdandCLAUDE.mdto avoid redundant discovery questions (SKILL.md). \n - Boundary markers: Instructions do not specify the use of delimiters or boundary markers when reading external project context. \n
- Capability inventory: The skill has the capability to write the final brief to
.design-context.mdin the project root (SKILL.md). \n - Sanitization: There is no mention of sanitization or validation of data read from local files before it is processed or used to generate the output artifact. \n- [SAFE]: The skill references a catalog of external media services and design tools including well-known providers such as NASA, Wikimedia, Pexels, and Gemini for specialized implementation needs (references/design-media-catalog.md).
Audit Metadata