aura-theme
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests free-form user descriptions to generate CSS configurations, creating a surface for potential injection.
- Ingestion points: User descriptions provided during the theme generation workflow defined in SKILL.md.
- Boundary markers: None explicitly defined to separate user input from the generation logic.
- Capability inventory: The skill writes new CSS files and modifies the existing styles.css file via @import statements as described in the workflow section of SKILL.md.
- Sanitization: The skill uses a predefined mapping table and reference files (references/property-values.md) to constrain output to valid Aura properties and values, mitigating the risk of arbitrary code injection.
- [EXTERNAL_DOWNLOADS]: The skill references font resources from a well-known service for web typography.
- Evidence: SKILL.md and reference CSS files (e.g., references/cozy-paper.css) include @import rules for https://fonts.googleapis.com.
- [SAFE]: The primary operations of the skill involve generating aesthetic configurations within standard project resource directories (e.g., /src/main/resources/META-INF/resources/). No suspicious command execution, exfiltration, or persistence patterns were detected.
Audit Metadata