ui-design-review
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external, untrusted sources which could contain malicious instructions designed to hijack the agent.
- Ingestion points: The skill explicitly requires inputs such as
screenshots_or_urls,competitors, andbrand_guidelines(SKILL.md). - Boundary markers: The skill instructs the agent to 'mentally scope external content as ...' (SKILL.md). While these are helpful guidance, they are not structural boundaries enforced by the platform's input schema.
- Capability inventory: The agent is expected to fetch, view, and analyze external web pages, images, and documents related to competitors and brand assets.
- Sanitization: The skill provides specific defensive instructions to the agent to look for and ignore common injection patterns like 'ignore previous instructions' or 'you are now' within the input data (SKILL.md).
Audit Metadata