clay-incident-runbook

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes bash scripts and diagnostic commands (curl, kubectl) designed for incident triage. These tools are used to verify service availability, test webhook connectivity, and inspect integration health in a production environment.
  • [DATA_EXFILTRATION]: No unauthorized data exfiltration patterns were detected. The scripts interact with the legitimate Clay API and user-configured webhooks to perform status checks using provided environment variables.
  • [CREDENTIALS_UNSAFE]: No hardcoded secrets or credentials are present. The skill correctly instructs the user or agent to use environment variables and Kubernetes secrets for sensitive information such as API keys.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 12:36 AM
Security Audit — agent-trust-hub — clay-incident-runbook