aws-cli

Pass

Audited by Gen Agent Trust Hub on Jun 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a comprehensive safety model that classifies AWS commands into 'Safe', 'Write', 'Destructive', and 'Forbidden' tiers. High-risk operations are deterministicly gated behind explicit user confirmation flows.
  • [COMMAND_EXECUTION]: The skill utilizes local shell scripts (scripts/aws_preflight.sh) and a Python classifier (scripts/aws_risk.py) to manage AWS environment configuration and risk assessment. These scripts operate locally on the user's machine to provide safety guardrails.
  • [EXTERNAL_DOWNLOADS]: The skill references official AWS documentation and API references from well-known domains such as awscli.amazonaws.com and docs.aws.amazon.com for self-healing and command verification purposes.
  • [SAFE]: Instructions explicitly forbid the automatic display of sensitive information such as access keys, session tokens, or secret values from AWS Secrets Manager and SSM Parameter Store.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 30, 2026, 11:31 AM
Security Audit — agent-trust-hub — aws-cli