brand-reel
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the extraction of branding data from external websites. This is identified as a vulnerability surface, but is assessed as safe given its restricted scope and primary role in the skill's functionality.\n
- Ingestion points: External website URLs provided for the
extract_brand_kittool as described in SKILL.md.\n - Boundary markers: None explicitly defined in the skill instructions.\n
- Capability inventory: Professional tools for video composition, media import, and rendering.\n
- Sanitization: The extraction tool is limited to retrieving visual metadata such as hex colors and font names, which minimizes the risk of executable content injection.
Audit Metadata