visual-modes

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install several third-party libraries using standard package managers (npm). These packages are well-known, established tools for animations, 3D rendering, and UI components, including Three.js, GSAP, tsparticles, and framer-motion.
  • [COMMAND_EXECUTION]: The skill uses bash to install project dependencies and the Edit/Write tools to implement visual enhancement patterns directly into the source code. These operations are consistent with the skill's primary purpose of UI development and interaction design.
  • [INDIRECT_PROMPT_INJECTION]: The skill identifies an attack surface by reading external project files (such as package.json, requirements.txt) and user-authored UI components to provide tailored design suggestions.
  • Ingestion points: Project configuration files and UI components matching patterns like **/canvas/** or **/design/** in supersaiyan.md and SKILL.md respectively.
  • Boundary markers: None explicitly defined for untrusted code input.
  • Capability inventory: File system modification (Edit/Write) and shell command execution (Bash) across all referenced files.
  • Sanitization: Not present; the skill relies on the agent's internal logic to filter and apply enhancements.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 12:41 PM
Security Audit — agent-trust-hub — visual-modes