theme-factory
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions for generating custom themes create an attack surface by consuming untrusted user data to drive agent behavior.
- Ingestion points: User-provided descriptions and inputs for custom themes as defined in the 'Create your Own Theme' section of SKILL.md.
- Boundary markers: The instructions lack delimiters or explicit warnings for the agent to ignore commands that may be embedded within the user's styling descriptions.
- Capability inventory: The skill directs the agent to apply styling to artifacts (slides, documents, HTML), which involves modifying files or calling presentation tools.
- Sanitization: No validation or sanitization routines are specified for the user input before it is used to generate and apply a theme.
Audit Metadata