The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install an external package globally using npm install -g figma-use. This package is not from a pre-verified trusted source.
[COMMAND_EXECUTION]: The skill relies on executing various figma-use commands to interact with the Figma design environment, including creating elements, modifying nodes, and exporting files. It also suggests running the Figma application with remote debugging enabled (--remote-debugging-port=9222).
[DYNAMIC_EXECUTION]: The figma-use render command processes JSX/TSX files and strings. This involves executing JavaScript/TypeScript logic at runtime to generate declarative design structures, which is a form of dynamic execution.
[INDIRECT_PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection by ingesting untrusted data from external design files and the Figma environment.
Ingestion points: The skill reads design data through figma-use node tree, figma-use node get, and figma-use query. It also processes external design templates via figma-use render and figma-use export jsx.
Boundary markers: There are no explicit instructions or delimiters shown to prevent the agent from following instructions embedded within the Figma design data or JSX templates.
Capability inventory: The skill allows for file writing (export), command execution (figma-use), and network fetching (create image --url).
Sanitization: No evidence of sanitization or validation of the ingested design content is provided in the skill definition.

figma-use-cli