incident-runbook-templates
Pass
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides numerous functional templates for system administration and incident mitigation, including commands for Kubernetes management (kubectl), database interaction (psql), and API calls (curl). These are intended for operational recovery and service troubleshooting.
- [PROMPT_INJECTION]: The 'Memory Protocol' section contains mandatory instructions that direct the agent to read from and write to specific local file paths (e.g., .claude/context/memory/learnings.md) to maintain state across execution sessions.
- [INDIRECT_PROMPT_INJECTION]: Ingestion points: The skill facilitates the processing of potentially untrusted data from application logs and database records via triage commands. Boundary markers: No specific delimiters or warnings are included in the templates to differentiate between logs and instructions. Capability inventory: The skill has access to Bash, Read, and Write tools, enabling it to execute commands based on interpreted content. Sanitization: There is no evidence of sanitization or filtering of external data before it is presented to the agent.
Audit Metadata