figma-implement-design

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill fetches design specifications and visual screenshots from Figma URLs and downloads graphical assets from a localhost endpoint managed by the Figma MCP server. These actions are restricted to the primary purpose of the skill and utilize trusted or local endpoints.\n- [PROMPT_INJECTION]: The skill ingests data from external Figma files which presents a potential surface for indirect prompt injection.\n
  • Ingestion points: Metadata and text content are fetched from external Figma URLs using the get_design_context tool as described in SKILL.md.\n
  • Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded content within the design data.\n
  • Capability inventory: The agent has the capability to generate and write application code to the user's repository based on the design context.\n
  • Sanitization: The skill does not specify procedures for sanitizing or escaping text content extracted from Figma layers before incorporating it into the generated source code.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 09:08 AM
Security Audit — agent-trust-hub — figma-implement-design