chaos-engineer

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides numerous example commands for executing chaos experiments using kubectl, aws cli, toxiproxy-cli, and pumba. These are standard DevOps operations intended for resilience testing in controlled environments.
  • [EXTERNAL_DOWNLOADS]: Instructions include downloading the litmus-operator manifest from litmuschaos.github.io and installing toxiproxy via brew. These references point to well-known community tools and official project repositories.
  • [CREDENTIALS_UNSAFE]: While the skill includes scripts for interacting with AWS and Gremlin APIs, it correctly demonstrates using environment variables and GitHub Action secrets (${{ secrets.AWS_ACCESS_KEY_ID }}, ${{ secrets.SLACK_WEBHOOK }}) rather than hardcoding credentials.
  • [PROMPT_INJECTION]: The skill contains a 'Safety Checklist' that explicitly instructs the agent to enforce constraints like blast radius caps and automated rollbacks, which aligns with safe operational practices for chaos engineering.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 04:39 PM
Security Audit — agent-trust-hub — chaos-engineer