sumeru-topic
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill performs expected content generation for novel planning. It writes reports to the current directory and stores metadata in
.sumeru/topic/. No suspicious network calls, credential access, or unauthorized command executions were found. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input for novel themes. 1. Ingestion points: User-provided keywords and genres. 2. Boundary markers: Absent. 3. Capability inventory: Local file write operations. 4. Sanitization: Absent. The risk is minimal given the restricted creative scope of the skill.
Audit Metadata