figma-implement-design

Pass

Audited by Gen Agent Trust Hub on Mar 25, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill represents an attack surface for indirect prompt injection because it ingests untrusted data from external Figma nodes and uses that information to generate code.\n
  • Ingestion points: Design data is retrieved via tool calls (get_design_context and get_metadata) to the Figma MCP server as described in SKILL.md.\n
  • Boundary markers: The skill relies on a structured multi-step workflow but does not define explicit delimiters or instructions to ignore embedded commands within the design data.\n
  • Capability inventory: The agent has the capability to write implementation code to the local file system based on the external design specifications.\n
  • Sanitization: There is no evidence of sanitization or validation of the text or metadata retrieved from Figma before it is used in the code generation process.\n- [SAFE]: The skill utilizes the official Figma MCP server (mcp.figma.com) to retrieve design context and assets. Figma is a well-known service, and using its official infrastructure for design translation is a legitimate behavior. Instructions for user setup, such as enabling the remote MCP client and logging in, are standard for the tool's intended use.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 25, 2026, 01:22 PM
Security Audit — agent-trust-hub — figma-implement-design