theme-factory
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (HIGH): The skill's primary function is to process and modify external artifacts (slides, docs, HTML landing pages). This creates a significant attack surface where malicious instructions embedded in the processed files could be executed by the agent.
- Ingestion points: Artifacts being styled (slides, docs, HTML) as mentioned in
SKILL.md. - Boundary markers: Absent. There are no instructions to ignore embedded commands or use delimiters when parsing artifacts.
- Capability inventory: File modification capability ("Apply the selected theme's colors and fonts to the deck/artifact").
- Sanitization: Absent. No validation or filtering is specified for the content within the artifacts being modified.
- [Prompt Injection] (MEDIUM): The 'Create your Own Theme' feature in
SKILL.mdallows the agent to generate new themes based on arbitrary 'provided inputs'. This is a direct vector for prompt injection where a user or data source could provide instructions disguised as a theme description to override agent behavior.
Recommendations
- AI detected serious security threats
Audit Metadata