figma-extract-design-system

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to perform directory operations and execute shell commands (cd, npm install, npm create) to initialize and configure the design system environment. It also utilizes specialized local tools (figma_ds_*) to perform file system analysis and code generation.
  • [EXTERNAL_DOWNLOADS]: Fetches package dependencies and project scaffolding tools from the NPM registry, which is a well-known and established service.
  • [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: Orchestrates the execution of npm create storybook@latest, which downloads and runs a remote initialization script from a well-known service to scaffold the workshop environment.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill ingests untrusted source code from target applications using the figma_ds_analyze and figma_ds_extract_component tools.
  • Boundary markers: The instructions lack explicit delimiters or safety warnings to prevent the agent from following instructions embedded within the analyzed source code.
  • Capability inventory: The agent has the capability to execute shell commands, write to the file system, and modify configuration files (e.g., main.js, preview.jsx).
  • Sanitization: There is no mention of sanitization or filtering of the extracted code before it is ported into the new design system package.
  • [DYNAMIC_EXECUTION]: The skill automatically modifies project configuration files and generates Storybook stories based on observed code patterns, representing a dynamic code generation workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 05:00 AM
Security Audit — agent-trust-hub — figma-extract-design-system