The Agent Skills Directory

[SAFE]: The skill exclusively utilizes Figma-specific MCP tools to read node properties and apply design styles. All operations are confined to the design context and do not involve shell access or general-purpose network capabilities.
[SAFE]: The instructions include logical safeguards, such as requiring a match between raw values and design system styles before applying links, and skipping component instances to prevent breaking intentional overrides.
[SAFE]: No evidence of prompt injection, obfuscation, or persistence mechanisms was detected in the skill instructions or metadata.

style-audit