redesign-existing-projects
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires reading and auditing an external project's codebase, which creates a vulnerability surface where malicious instructions embedded in the codebase could influence the agent's behavior. • Ingestion points: Reads all files in the project codebase during the 'Scan' phase to identify styling methods and patterns. • Capability inventory: Modifies source code files, styling configurations, and potentially adds dependencies based on the audit. • Boundary markers: No specific delimiters are defined to separate the external codebase content from the skill's own instructions. • Sanitization: No explicit validation or filtering of the ingested codebase content is specified.
- [EXTERNAL_DOWNLOADS]: The skill suggests fetching placeholder imagery from the well-known service picsum.photos when local assets are unavailable.
Audit Metadata