design-dna-extractor
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external URLs and images during the 'Analyze' phase. It extracts high-level concepts like 'brand-voice' and 'messaging-approach' which involve natural language processing. * Ingestion points: External URLs and uploaded image files (screenshots) processed by the agent's vision capabilities. * Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the documentation for the data extraction phase. * Capability inventory: The skill has network access to fetch external data and the ability to generate executable web code (HTML, CSS, and JavaScript). * Sanitization: No validation or sanitization of the extracted content is described before it is used in the 'Generate' phase.
- [EXTERNAL_DOWNLOADS]: The skill performs network operations to analyze visual properties of external websites provided by the user.
- [COMMAND_EXECUTION]: The documentation includes standard installation instructions using
npxandgit cloneto retrieve code from external repositories (e.g.,github.com/zanwei/design-dna.git).
Audit Metadata