troubleshoot-clickhouse

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides structured guidance and MCP query templates for diagnosing ClickHouse failure modes such as merge debt spirals and memory exhaustion.
  • [SAFE]: All referenced system paths (e.g., /proc filesystem) and network endpoints are documented as standard data sources for the Netdata collector, not as targets for exfiltration or direct agent execution.
  • [SAFE]: No obfuscation, hardcoded credentials, or persistence mechanisms were detected. The skill adheres to the expected behavior of a monitoring and troubleshooting extension.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 11:23 AM
Security Audit — agent-trust-hub — troubleshoot-clickhouse