chaos-engineering-on-aws
Pass
Audited by Gen Agent Trust Hub on Apr 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes several shell scripts (scripts/monitor.sh, scripts/log-collector.sh, scripts/setup-prerequisites.sh) that execute AWS CLI and kubectl commands. These are used to create experiment templates, start experiments, and gather metrics or logs, which is consistent with the skill's primary function.
- [EXTERNAL_DOWNLOADS]: The skill references external resources for installation and setup, including cloning repositories from GitHub (aws-samples) and installing Chaos Mesh via its official Helm repository (charts.chaos-mesh.org). It also utilizes MCP servers from the awslabs organization and provides an optional reference to a third-party MCP server (RadiumGu/Chaosmesh-MCP).
- [PROMPT_INJECTION]: The skill features an indirect prompt injection surface (Category 8). 1. Ingestion points: reads risk inventories and resilience reports in SKILL_EN.md Steps 1-3. 2. Boundary markers: absent from interpolation logic. 3. Capability inventory: includes infrastructure manipulation via FIS and Kubernetes tools. 4. Sanitization: no escaping or validation of report data is described.
Audit Metadata