brand-guidelines
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided artifacts to apply brand styling, creating a surface for indirect prompt injection where instructions hidden within user data could be interpreted by the agent.\n
- Ingestion points: Processes any sort of artifact provided by the user (as described in SKILL.md).\n
- Boundary markers: No specific delimiters or instructions to ignore embedded content are defined within the skill's guidelines.\n
- Capability inventory: The skill possesses the capability to modify documents and visual assets, potentially using file-writing tools and libraries like python-pptx.\n
- Sanitization: There is no evidence of sanitization, filtering, or validation of the input artifact content before processing.\n- [EXTERNAL_DOWNLOADS]: The skill's technical documentation references the use of the python-pptx library for document manipulation.\n
- Evidence: SKILL.md mentions that color application is handled via python-pptx's RGBColor class.
Audit Metadata