The Agent Skills Directory

[SAFE]: The skill focuses on legitimate integration with the Figma API via MCP tools. No malicious patterns, obfuscation, or unauthorized data access were detected.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) as it processes untrusted data from external Figma files.
Ingestion points: Design context and metadata are fetched from Figma URLs in SKILL.md and references/figma-skill.md.
Capability inventory: The skill can modify Figma canvases (use_figma), create files (create_new_file), and update code mappings (send_code_connect_mappings).
Boundary markers: The skill lack specific boundary markers for Figma content but includes a procedural guardrail in SKILL.md requiring user confirmation for high-risk destructive edits.
Sanitization: No explicit sanitization or instruction filtering is implemented for the data retrieved from Figma.

figma-skill