The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill connects to the official Figma MCP server at https://mcp.figma.com/mcp to fetch design context, metadata, and screenshots. This is a standard integration with a well-known service.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted design data (node names, comments, and metadata) from external Figma files that could theoretically contain malicious instructions.
Ingestion points: Design data is fetched via the get_design_context tool in SKILL.md and references/figma-tools-and-prompts.md.
Boundary markers: No specific delimiters are used to separate design data from instructional prompts.
Capability inventory: The skill allows the agent to generate and implement React and Tailwind CSS code based on the design input.
Sanitization: The skill does not explicitly describe sanitization or filtering of the data returned by the Figma MCP server.
[PERSISTENCE]: Documentation in references/figma-mcp-config.md provides instructions for the manual addition of the FIGMA_OAUTH_TOKEN to user shell profiles (e.g., ~/.bashrc, ~/.zshrc) to persist the environment variable across sessions. This is a standard and documented setup procedure for developers.

figma