The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from the Figma REST API and external connectors, which could contain malicious instructions.
Ingestion points: Node trees, variables, and metadata fetched from figma.com URLs or the Figma API in multiple phases (Asset, Tokens, Component).
Boundary markers: The instructions do not define explicit delimiters or warnings for the agent to ignore instructions embedded in the design data.
Capability inventory: The skill can execute shell commands via npx and a local Node.js script (scripts/verify-motion.mjs) which calls ffmpeg and ffprobe.
Sanitization: The documentation notes that SVG assets are sanitized during the import process.
[EXTERNAL_DOWNLOADS]: The skill utilizes npx to fetch and update the hyperframes CLI tool from the npm registry as part of its setup and maintenance workflow.
[COMMAND_EXECUTION]: A bundled script, scripts/verify-motion.mjs, uses execFileSync and spawnSync to perform video analysis. While it processes file paths provided as arguments, it uses secure argument passing (arrays) to prevent shell-based command injection.

figma