brandkit
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely composed of conceptual guidance and artistic principles for generating brand identity imagery. It does not include scripts, executables, or command-line instructions.
- [DATA_EXPOSURE_AND_EXFILTRATION]: No access to local files, environment variables, or credentials was detected. The skill does not perform network requests.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a prompt template for generating image descriptions based on user-provided brand names and categories. 1. Ingestion points: The internal PROMPT TEMPLATE accepts [BRAND NAME] and [category] variables from user input. 2. Boundary markers: None. 3. Capability inventory: The skill is limited to image generation and does not have capabilities for file operations or remote execution. 4. Sanitization: None specified, but the limited output format (images) prevents the execution of malicious instructions.
- [METADATA_POISONING]: The metadata accurately describes the skill's purpose as an image generation tool and contains no deceptive instructions.
Audit Metadata