aws-resilience-modeling
Pass
Audited by Gen Agent Trust Hub on Apr 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to executeawsandghCLI commands to perform resource discovery, configuration audits, and repository management. - [COMMAND_EXECUTION]: Runs local Python scripts (
generate-html-report.py) and dynamic Python code via shell heredocs to assemble and format assessment data into reports. - [EXTERNAL_DOWNLOADS]: Fetches interactive visualization libraries (Chart.js v4.4.0 and Mermaid v10) from well-known CDNs (jsdelivr.net) within the generated HTML report template.
- [EXTERNAL_DOWNLOADS]: Recommends the installation and execution of official Model Context Protocol (MCP) servers from the
awslabsorganization (e.g.,awslabs.aws-api-mcp-server) using theuvxpackage runner. - [DATA_EXFILTRATION]: Accesses architectural metadata, resource configurations, and CloudWatch monitoring data from the user's AWS account to perform resilience analysis. While this involves reading sensitive infrastructure data, it is the primary intended purpose of the skill and originates from a trusted vendor.
- [PROMPT_INJECTION]: As the skill ingests and processes external architectural documentation and AWS resource metadata (such as tags and descriptions), it possesses a surface for indirect prompt injection (Category 8).
Audit Metadata