incident-runbook-templates
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes numerous templates with functional shell commands for
kubectl,psql, andcurl. These are intended for infrastructure troubleshooting and incident mitigation tasks such as scaling services, rolling back deployments, and managing database connections. - [PROMPT_INJECTION]: The runbook templates involve instructions for the agent to analyze application logs (e.g.,
kubectl logs -n payments -l app=payment-service --tail=500). This creates a surface for indirect prompt injection where untrusted data embedded in logs could influence the agent's logic or downstream commands. - Ingestion points: Log analysis command examples in
SKILL.md. - Boundary markers: Absent. No specific delimiters or warnings to ignore embedded instructions are provided in the templates.
- Capability inventory: The skill has access to significant system capabilities including
kubectl(scale, rollout, networking policies),psql(query termination), andcurl(internal API interaction). - Sanitization: The skill does not provide methods for sanitizing or escaping log content before it is processed by the agent.
Audit Metadata