design-system-foundations-toolkit

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill defines workflows for ingesting and auditing external files such as design tokens and source code.
  • Ingestion points: Processes design tokens in JSON (DTCG format) and component source files (TSX, CSS, JSX) within the project structure.
  • Boundary markers: The skill relies on structured technical formats but does not implement specific delimiters to prevent the agent from potentially following instructions embedded in these data files.
  • Capability inventory: Includes shell command execution instructions (grep) for auditing and JavaScript snippets for runtime style evaluation.
  • Sanitization: No explicit filtering or content-based validation is implemented for the data processed by the toolkit.
  • [DYNAMIC_EXECUTION]: The toolkit provides templates for build configurations (e.g., style-dictionary.config.js) and testing scripts (Playwright) that are intended to be generated and executed within the development environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 05:48 PM
Security Audit — agent-trust-hub — design-system-foundations-toolkit