dark-mode
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides comprehensive guidelines for dark mode implementation, including CSS tokens, hierarchy, and text contrast rules. All provided code snippets (CSS, JS, HTML) are standard for client-side theme management and present no security risk to the agent or the execution environment.
- [INDIRECT_PROMPT_INJECTION]: The skill acknowledges external data ingestion (Figma URLs) in the design workflow. However, it lacks dangerous capabilities such as file writes, network exfiltration, or shell execution, neutralizing the risk. Evidence chain: 1. Ingestion points: User-provided Figma URLs (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: None detected across the skill scripts. 4. Sanitization: Absent.
Audit Metadata