design-md-format
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references the installation and usage of the
@google/design.mdpackage from the npm registry. This package originates from a trusted organization. - [COMMAND_EXECUTION]: The documentation provides examples for executing CLI commands such as
npx @google/design.md lintanddesignmd export. These commands are used for validating design systems and converting tokens into configuration files for Tailwind CSS or W3C Design Tokens. - [INDIRECT_PROMPT_INJECTION]: The skill establishes a workflow for processing
DESIGN.mdfiles, which may contain data from various project sources. - Ingestion points: The agent is tasked with creating, validating, and linting
DESIGN.mdfiles that contain YAML tokens and Markdown prose (SKILL.md). - Boundary markers: The specification utilizes structured YAML frontmatter to encapsulate design tokens, providing a clear boundary between machine-readable data and human-readable rationale.
- Capability inventory: Capabilities include reading files and executing shell commands for linting, diffing, and exporting data (SKILL.md).
- Sanitization: The provided linter automatically validates structural correctness, resolves token references, and checks WCAG contrast ratios, which serves as a security control for the data being processed.
Audit Metadata