uniapp
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines templates for rendering external user data (e.g., nicknames, emails) which creates a potential surface for indirect prompt injection if the ingested data contains malicious instructions targeting the agent.
- Ingestion points: The UserCard component and list templates in SKILL.md process UserVO data objects retrieved from UserAPI.
- Boundary markers: The provided templates do not include specific boundary markers or 'ignore instructions' warnings for the interpolated data fields.
- Capability inventory: The code templates include capabilities for navigation (uni.navigateTo), modal dialogs (uni.showModal), and API calls (UserAPI.deleteById).
- Sanitization: No explicit sanitization or escaping logic is mandated in the standards for the data interpolation steps.
Audit Metadata