figma-implement-motion

Pass

Audited by Gen Agent Trust Hub on Nov 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to download assets (images and SVGs) using curl -L from the Figma MCP server. These downloads target temporary, authenticated asset URLs generated by the official Figma integration.
  • [COMMAND_EXECUTION]: The skill prescribes the use of shell commands (specifically curl) to retrieve remote assets into the local project directory. These commands are limited to downloading media files and do not involve pipe-to-shell or other high-risk execution patterns.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external Figma files, creating an attack surface for indirect prompt injection from malicious Figma file content.
  • Ingestion points: Figma URLs provided by the user and JSON responses from get_design_context and get_motion_context tools.
  • Boundary markers: The instructions explicitly define get_motion_context as the "authoritative motion tool" and the "source of truth," establishing clear data priorities.
  • Capability inventory: The skill has capabilities for file system writes (writing implementation code) and network downloads via curl.
  • Sanitization: The skill emphasizes structural merging of code snippets and design data rather than raw text interpolation, reducing the risk of accidental instruction obedience from data.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 17, 2026, 03:12 PM