theme-factory
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates entirely on local theme definitions and stylistic instructions. It lacks capabilities for network communication, system command execution, or sensitive file access.- [INDIRECT_PROMPT_INJECTION]: The skill allows users to provide descriptions for generating custom themes, which creates a data ingestion point. However, this is assessed as safe because the resulting data is restricted to visual styling parameters (hex codes, fonts) and is not processed by high-privilege tools.
- Ingestion points: User-supplied inputs for creating custom themes in SKILL.md.
- Boundary markers: No explicit delimiters are used to wrap user input.
- Capability inventory: Reading theme files and applying stylistic changes to user-specified artifacts.
- Sanitization: The skill does not perform specific validation on user input beyond thematic interpretation.
Audit Metadata