aws-resilience-modeling
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches Model Context Protocol (MCP) servers from official AWS labs repositories (
awslabs) and frontend libraries for visualization from established Content Delivery Networks (jsdelivr). These are trusted and well-known sources within the cloud development ecosystem. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external architectural documentation and Infrastructure-as-Code (Terraform/CloudFormation) files provided by the user. This represents a standard ingestion surface necessary for its primary function of resilience modeling.
- [COMMAND_EXECUTION]: The skill uses the AWS CLI for basic identity verification via
aws sts get-caller-identityand interacts with MCP servers for automated environment scanning. The instructions explicitly mandate read-only access for these operations, demonstrating a security-conscious design. - [DYNAMIC_EXECUTION]: The skill includes a local Python utility script to generate interactive HTML reports. The instructions guide the agent to dynamically load this module from the skill's local directory to process assessment data and populate the report template.
Audit Metadata