theme-factory
Warn
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: MEDIUMPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits metadata poisoning by claiming 'Anthropic' as the author in the SKILL.md frontmatter, which contradicts the provided author identity 'zocomputer'. This deceptive practice can mislead users into misplacing trust in the skill's origin and safety.
- [PROMPT_INJECTION]: The 'Create your Own Theme' functionality creates a surface for indirect prompt injection by instructing the agent to generate new styling instructions based on unvalidated user descriptions and subsequently apply them to artifacts.
- Ingestion points: User-supplied theme descriptions and names within the 'Create your Own Theme' workflow in SKILL.md.
- Boundary markers: Absent; there are no instructions to ignore embedded commands or treat user input as data only.
- Capability inventory: The agent is authorized to modify artifacts (slides, docs, HTML) by applying generated themes.
- Sanitization: Absent; the skill does not specify any filtering or validation for the generated theme content before application.
Audit Metadata