incident-runbook-templates
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill contains templates with shell commands for Kubernetes cluster management, database administration, and system resource checks.
- [EXTERNAL_DOWNLOADS]: References official status pages and documentation from well-known technology providers such as Stripe, Google, and PagerDuty.
- [REMOTE_CODE_EXECUTION]: Mentions local scripts located in the
./scripts/folder for executing smoke tests and rollbacks. - [DATA_EXFILTRATION]: Includes network operations via
curlto interact with monitoring systems and internal feature flag APIs. - [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection. Ingestion points: reads untrusted data via
kubectl logs(SKILL.md). Boundary markers: none present to delimit log content. Capability inventory: includes subprocess execution for cluster management, database queries, and network requests (kubectl,psql,curlin SKILL.md). Sanitization: no evidence of escaping or validation of log content before processing.
Audit Metadata