The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill requires the agent to read and analyze local source code files (TypeScript, React, Kotlin, Swift, and CSS) to identify UI patterns and design system mappings. This ingestion of potentially untrusted data represents an attack surface for indirect prompt injection.
Ingestion points: Source code files (*.tsx, *.ts, *.kt, *.swift, *.figma.ts), global stylesheets, and SVG assets are read from the local project to guide design generation.
Boundary markers: The instructions do not specify the use of delimiters or boundary markers when the agent processes these files.
Capability inventory: The agent uses the use_figma tool to execute generated JavaScript in the Figma environment and can invoke tools to search design systems and create files.
Sanitization: No specific sanitization or validation of the content extracted from source code (such as SVG strings or CSS values) is described before it is interpolated into Figma scripts.
[COMMAND_EXECUTION]: The agent is instructed to use shell-based tools (glob, grep) to search the user's codebase for design system configuration files and UI components.
[DYNAMIC_EXECUTION]: The skill's primary function is the generation and execution of JavaScript scripts within the Figma plugin environment via the use_figma tool.

figma-generate-design