open-design
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download binaries from open-design.ai or clone source code from the official GitHub repository at github.com/nexu-io/open-design. These links represent the vendor's primary distribution channels.\n- [INDIRECT_PROMPT_INJECTION]: The application processes user-provided natural language briefs and instructions to generate functional code and previews, creating a potential surface for malicious instructions to be embedded in the processed data.\n
- Ingestion points: Design briefs and descriptions provided by the user in plain language (documented in SKILL.md).\n
- Boundary markers: No delimiters or specific security instructions for handling embedded commands are mentioned in the provided documentation.\n
- Capability inventory: The platform generates web prototypes, landing pages, and provides an MCP server to interact with other agents.\n
- Sanitization: No sanitization or validation mechanisms for the input briefs are specified in the stub configuration.
Audit Metadata