brand-onboarding
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution: The skill utilizes
uv run design-agentandshasumto initialize brand packages, verify fonts, and generate CSS. These commands are integral to the workflow and facilitate the automation of design tasks. - Data Ingestion Surface: The process involves reading information from external sources like PDFs, websites, and design files. While processing untrusted data can introduce a surface for indirect instructions, the skill emphasizes validation using specific subcommands and manual review by a person.
- Sensitive File Handling: It provides clear instructions for managing licensed fonts in a separate directory (
private-fonts/) and explicitly warns against committing them to version control, which is a positive practice for protecting intellectual property.
Audit Metadata