brandkit-build
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill generates an HTML template (
assets/brand-helmet.html.template) that includes a stylesheet link to Google Fonts (https://fonts.googleapis.com/css2). This is a well-known service used for the intended purpose of rendering typography. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (a "brand interview") to generate binding instructions for the AI agent (
CLAUDE.mdandBRAND.md). - Ingestion points: Reads a raw brand interview file (referenced in
SKILL.md). - Boundary markers: None explicitly defined for the input data; the output is structured into Markdown tables and CSS blocks.
- Capability inventory: The skill performs file creation and content formatting. It does not possess network-write or command-execution capabilities.
- Sanitization: There is no evidence of sanitization or escaping of the input text before it is written into the instruction files.
- [COMMAND_EXECUTION]: The skill mentions using Git for version control, but it does not automatically execute shell commands or require administrative privileges.
Audit Metadata