agent-observability
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains purely instructional content focused on agent observability best practices. No executable scripts or suspicious commands are present.
- [DATA_EXPOSURE]: The instructions explicitly recommend redacting user data at the collection boundary to prevent accidental exposure within traces, demonstrating a security-first approach to telemetry.
- [NO_CODE]: The skill consists entirely of markdown documentation without any code, package dependencies, or automated execution patterns that could introduce risk.
Audit Metadata