theme-factory
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious code, credential exposure, or persistence mechanisms were found. The skill operates on static markdown data.
- [NO_CODE]: There are no Python or JavaScript files provided in the skill package; the primary logic is handled via natural language instructions for the agent.
- [PROMPT_INJECTION]: The 'Create your Own Theme' functionality in SKILL.md presents a surface for indirect prompt injection via user input interpolation.
- Ingestion points: User-supplied descriptions for custom themes.
- Boundary markers: Absent; there are no delimiters or explicit instructions to ignore embedded commands in the user-provided text.
- Capability inventory: The agent is instructed to read theme files and apply styles to artifacts (such as slide decks or documents), which involves file modification capabilities.
- Sanitization: No validation or sanitization of the user input is specified in the application process.
Audit Metadata