redesign-existing-projects
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data (existing project codebases) and then perform file modifications based on that analysis. This creates a potential surface for indirect prompt injection if the source code contains malicious instructions.
- Ingestion points: The 'Scan' step explicitly instructs the agent to read the project's codebase to identify frameworks, styling methods, and patterns (SKILL.md).
- Boundary markers: The instructions do not define specific delimiters or security warnings to treat the ingested codebase as untrusted data.
- Capability inventory: The skill has the capability to write and modify files during the 'Fix' stage to apply design upgrades (SKILL.md).
- Sanitization: There are no explicit sanitization or validation steps mentioned for the content read from the codebase before it is processed by the agent.
- [EXTERNAL_DOWNLOADS]: The skill recommends using
https://picsum.photos/seed/{name}/1920/1080for placeholder images. This targets a well-known and standard service for web development placeholders and is considered a safe operational reference.
Audit Metadata