kove-weekly-review
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-generated content from journal notes which represents a surface for indirect prompt injection. 1. Ingestion points: Journal text is ingested via the
get_daystool inSKILL.md. 2. Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for the retrieved data. 3. Capability inventory: The skill can read various health metrics and has write access viaadd_insights_card. 4. Sanitization: No sanitization or validation of the journal content is mentioned.
Audit Metadata