security-digest
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates as a purely analytical tool, processing event data through established correlation rules (CORR-01 through CORR-05) and a severity model based on time, location, and frequency.
- [SAFE]: It emphasizes privacy and data integrity by instructing the agent to never request secrets and to report explicitly on the status of all data sources (available, unavailable, partial, or capped).
- [SAFE]: The skill uses parallel batch processing for efficiency but includes strict budget limits (maximum of six calls or 60 seconds) to prevent excessive resource consumption or data scraping.
- [SAFE]: Correlation logic is deterministic and requires pre-configured metadata (such as camera-to-door mappings), preventing the agent from fabricating security alerts based on timing or proximity alone.
Audit Metadata