minimal-web-baas-demo
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to process and interpret user-provided requirements (e.g., requests for 'message boards' or 'Kanban') to generate code, database schemas, and configuration. As with any agent skill that translates natural language to executable logic or infrastructure, there is an inherent surface where adversarial user input could attempt to manipulate the resulting application structure.
- [EXTERNAL_DOWNLOADS]: The skill's workflow includes fetching project templates via a
downloadTemplatetool and managing dependencies through standard package managers (npm install,pnpm install). These operations target the CloudBase vendor's ecosystem and official package registries, which are standard for the intended development use case.
Audit Metadata