The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external Figma URLs through tools like get_design_context and get_metadata.
Ingestion points: Design data is ingested from Figma nodes specified in URLs provided by users or external sources (SKILL.md).
Boundary markers: The skill does not define explicit delimiters or instructions to ignore embedded content within the design data.
Capability inventory: The skill provides instructions for the agent to write files to the local disk (e.g., decoding base64 assets and saving them to public/assets/ as shown in SKILL.md).
Sanitization: There is no evidence of sanitization or validation for node names or design content before they are used in file system operations, creating a vector for path traversal if node names are manipulated.
[EXTERNAL_DOWNLOADS]: The skill connects to a remote MCP server at https://mcp.figma.com/mcp and provides links to extension guides and code repositories on Figma's official GitHub organization.
[COMMAND_EXECUTION]: The documentation includes CLI commands for installing plugins (claude plugin install) and configuring MCP servers (claude mcp add). It also provides an integration script example using child_process.exec to run agent commands.

figma-mcp-server