runbooks

Pass

Audited by Gen Agent Trust Hub on Aug 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely documentation-based, providing best practices for engineers writing incident recovery procedures.
  • [SAFE]: No executable scripts, subprocess calls, or automated command-line tools are provided within the skill.
  • [SAFE]: The provided code examples, such as metric checks using curl, are illustrative samples for inclusion in documentation and are not executed by the skill itself.
  • [SAFE]: Mentions of third-party platforms like Slack and PagerDuty are contextually appropriate references for alerting workflows and do not involve credential handling or API interaction.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 4, 2026, 01:49 AM
Security Audit — agent-trust-hub — runbooks