ui-design-review
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill package is composed entirely of Markdown documentation, a license file, and a static YAML configuration for the agent interface. No scripts (.py, .js, .sh), binaries, or package manifests were detected, precluding any direct code-based attacks.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external design artifacts, including mockups and product specifications. While this creates a theoretical ingestion surface for indirect prompt injection, the skill does not define or request any tools (e.g., shell access, file writes, or network requests) that could be leveraged to execute malicious instructions contained within those artifacts.
- [SAFE]: The instructions reinforce agent safety by explicitly forbidding the agent from making autonomous product decisions, self-authorizing fixes, or overriding product authority. The focus is strictly on reporting discrepancies back to a human-in-the-loop or a calling process.
Audit Metadata