startup-idea
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is purely instructional and contains no executable code, shell scripts, or binaries. It functions as a complex system prompt for business analysis.
- [NO_CODE]: No external dependencies, package managers (npm/pip), or remote downloads are present in the skill files.
- [PROMPT_INJECTION]: The skill ingests untrusted user data in the form of startup ideas. While it lacks explicit delimiters for this input, its systematic, multi-step analytical process effectively constrains the agent's behavior. The absence of tools or file-write capabilities prevents any potential injection from causing harm to the system.
- [SAFE]: Metadata and repository links are consistent with the skill's purpose. A minor naming variation between the author context and repository URL was noted but carries no security risk as the URL is purely informational and no code is fetched from it.
Audit Metadata