redesign-existing-projects
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface by reading untrusted user codebase content during the 'Scan' phase and possessing the capability to modify those files in the 'Fix' phase. This is an inherent risk for any skill that automates code refactoring.
- Ingestion points: Scan and Diagnose steps in SKILL.md.
- Boundary markers: Not present; the instructions do not include specific delimiters or warnings to ignore instructions found within the scanned codebase.
- Capability inventory: The agent is instructed to modify the user's styling and layout files.
- Sanitization: Not present; there is no logic described for escaping or validating the content of the ingested files before processing.
- [EXTERNAL_DOWNLOADS]: The skill recommends using
picsum.photosas a source for placeholder imagery. This is a well-known and widely used development utility for UI prototyping.
Audit Metadata