brandkit
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied brand names, metaphors, and visual assets which are interpolated into prompt templates, creating a potential surface for indirect prompt injection.
- Ingestion points: Untrusted data enters the agent context through the BRAND INTAKE section and user-supplied asset references.
- Boundary markers: The skill uses simple bracketed placeholders in its PROMPT TEMPLATE but does not include robust delimiters or explicit instructions to the agent to disregard commands embedded within the user data.
- Capability inventory: The skill is designed for visual prompt construction and art direction; it does not contain or reference tools for file system access, network communication, or shell command execution.
- Sanitization: No input validation, escaping, or filtering of user-supplied data is described in the instructions.
- [NO_CODE]: The skill consists entirely of instructional Markdown content and does not include any executable scripts, binaries, or library dependencies that would grant the agent additional system permissions.
Audit Metadata