The Agent Skills Directory

[SAFE]: The skill provides legitimate, high-quality documentation for managing design systems. It focuses on architecture, token formats, and development workflows without any malicious intent or security bypasses detected.
[EXTERNAL_DOWNLOADS]: The skill references the installation and use of official packages from established services, including @figma/code-connect, @anthropic/figma-mcp-server (from a well-known service provider), and standard development tools like Mitosis and Storybook. All downloads target official registries and trusted repositories for their respective domains.
[COMMAND_EXECUTION]: Provides standard CLI examples for common development tasks, such as starting a Storybook server (npx storybook dev), compiling cross-framework components (npx mitosis compile), and building design tokens (npx style-dictionary build). These commands are standard for the technologies described.
[CREDENTIALS_UNSAFE]: Correctly demonstrates the use of placeholders and environment variables for sensitive data, such as FIGMA_ACCESS_TOKEN: "<token>" and $FIGMA_TOKEN, which is a recommended security practice for handling API keys in documentation.
[INDIRECT_PROMPT_INJECTION]: The skill includes an attack surface for processing design-related data (DTCG JSON tokens and Figma design context). However, these inputs are processed by established build tools and document generation systems, posing no unique security risk in this context.

design-system