theme-factory
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because its primary function is to ingest, process, and modify untrusted user artifacts such as presentation decks, documents, and HTML landing pages. A malicious artifact could contain embedded instructions that the agent might inadvertently follow while performing the styling tasks.
- Ingestion points: Untrusted user artifacts (slides, docs, reports, HTML pages) as specified in the SKILL.md and used throughout the application process.
- Boundary markers: Absent. The instructions do not provide delimiters or "ignore embedded instructions" warnings to the agent when processing the external content of the artifacts.
- Capability inventory: The agent has the capability to read artifact contents and perform write/modification operations to apply themes, which could be exploited to alter the artifact's functional content or inject malicious scripts into HTML outputs.
- Sanitization: Absent. No validation, escaping, or filtering is defined for the inputs derived from the artifacts or the user-provided descriptions for custom theme generation.
Audit Metadata