taste-redesign-existing-projects

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill operates by scanning and modifying the user's codebase. This creates a potential surface for indirect prompt injection if the codebase contains malicious instructions. However, this behavior is essential to the skill's primary function as a redesign tool.
  • Ingestion points: Reads framework, styling, and design patterns from the existing codebase (SKILL.md).
  • Boundary markers: No specific delimiters are defined to separate code from instructions during processing.
  • Capability inventory: Authorized to modify CSS and HTML files to apply design upgrades (SKILL.md).
  • Sanitization: No explicit validation or filtering of codebase content is mentioned.
  • [EXTERNAL_DOWNLOADS]: The instructions suggest using picsum.photos as a source for high-quality placeholder images. This is a well-known, safe, and standard service for developers and does not pose a security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 07:10 AM
Security Audit — agent-trust-hub — taste-redesign-existing-projects