The Agent Skills Directory

[PERSISTENCE_MECHANISMS]: The reference documentation in references/figma-mcp-config.md instructs users to persist the FIGMA_OAUTH_TOKEN by adding export commands to shell profiles like ~/.zshrc or ~/.bashrc.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill manages a sensitive FIGMA_OAUTH_TOKEN and connects to a remote Figma MCP server at https://mcp.figma.com/mcp, which is recognized as a well-known and trusted service.
[INDIRECT_PROMPT_INJECTION]: The skill retrieves design context and metadata from the external Figma API to generate production code, creating a surface where malicious instructions could potentially be embedded in design nodes.
Ingestion points: Untrusted data enters the agent context via the get_design_context and get_metadata tools from mcp.figma.com (noted in SKILL.md and references/figma-tools-and-prompts.md).
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are used when processing the external design data.
Capability inventory: The agent has the capability to generate and implement code changes in the project repository based on the fetched data.
Sanitization: There is no evidence of sanitization or validation of the retrieved design data before it is used to drive code generation.

figma