redesign-existing-projects
Pass
Audited by Gen Agent Trust Hub on Jul 9, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill's workflow involves reading and modifying external codebase files, which introduces a surface for indirect prompt injection. If a processed file contains malicious instructions (e.g., in code comments or text content), the agent might inadvertently execute them while performing redesign tasks.\n
- Ingestion points: Reads the framework, styling method, and design patterns from the user's codebase (SKILL.md).\n
- Boundary markers: Absent; the instructions do not specify how to distinguish between legitimate code and untrusted instructions within the project files.\n
- Capability inventory: Modifications to the tech stack, CSS, and application code (SKILL.md).\n
- Sanitization: No validation or sanitization of codebase content is mentioned before processing.\n- [EXTERNAL_DOWNLOADS]: Suggests the use of https://picsum.photos/ for background imagery and assets when original content is unavailable. This is a common and reputable placeholder image service.
Audit Metadata