figma-make-project-converter

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute local JavaScript files using Node.js, specifically scripts/figma-make-converter.mjs and scripts/check-app-ready.mjs. These scripts perform file system operations, code transformations, and project analysis.
  • [EXTERNAL_DOWNLOADS]: The skill uses pnpm add <package-name> to install necessary runtime dependencies identified during the conversion process from the NPM registry.
  • [PROMPT_INJECTION]: The skill is vulnerable to Indirect Prompt Injection (Category 8). It processes arbitrary files from a user-provided directory (<figma-make-project-dir>), including documentation (DESIGN_SYSTEM_GUIDE.md) and code. The results of this analysis are written to task files (e.g., .figma-make-tasks.md) which the agent is then explicitly told to "check carefully" and follow to complete the conversion.
  • Ingestion points: Files within the user-provided <figma-make-project-dir> are read and processed.
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands within the ingested content are defined.
  • Capability inventory: The agent can execute shell commands (node, pnpm) and modify the local file system within the src/ directory.
  • Sanitization: There is no mention of sanitizing or escaping content from the source project before it is included in the AI-facing task reports.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 06:31 AM
Security Audit — agent-trust-hub — figma-make-project-converter