The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill connects to the official Figma MCP server at https://mcp.figma.com/mcp and downloads design assets. Figma is a well-known service, and these connections are necessary for the skill's primary function.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests untrusted design data and metadata from Figma design files.
Ingestion points: Data is fetched from Figma nodes and metadata using tools like get_design_context and get_metadata (references/figma-tools-and-prompts.md).
Boundary markers: The instructions in SKILL.md explicitly direct the agent to prioritize existing project code conventions and design tokens over the raw output generated from Figma.
Capability inventory: The agent uses the ingested design data to generate React/Tailwind code and download design assets.
Sanitization: There is no explicit description of sanitizing or escaping the Figma node content before processing.
[CREDENTIALS_UNSAFE]: The documentation in references/figma-mcp-config.md provides instructions for configuring a FIGMA_OAUTH_TOKEN as an environment variable. This is a standard and recommended practice for secret management in development environments.

figma