monitoring-expert
Pass
Audited by Gen Agent Trust Hub on Aug 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill includes clear security guidelines in its constraints, such as the requirement to redact sensitive data (PII, tokens, passwords) from logs.
- [CREDENTIALS_UNSAFE]: The configuration examples for Alertmanager and other services use secure placeholders like
'your-key'and'https://hooks.slack.com/...', preventing accidental exposure of real credentials. - [COMMAND_EXECUTION]: The application profiling documentation mentions common performance tools such as
clinic.js,py-spy, andpprof. These commands are standard for debugging and optimizing applications and do not involve suspicious remote execution patterns. - [EXTERNAL_DOWNLOADS]: The skill references several well-known libraries and tools from official package registries (NPM, PyPI) and established open-source projects (k6, Prometheus, OpenTelemetry). These are legitimate dependencies for an observability professional.
Audit Metadata