aws-cost-cleanup

Pass

Audited by Gen Agent Trust Hub on Oct 27, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from the AWS environment, including resource IDs, tags, and status information, via aws ec2 describe-* and boto3 calls in SKILL.md. The agent maintains capabilities to delete or modify cloud resources through aws ec2 delete-volume, aws ec2 delete-snapshot, and aws ec2 release-address. Documentation includes boundary markers such as a 'Safety Checklist' and 'Dry-run first' instructions, but the provided scripts do not demonstrate explicit sanitization of resource metadata before processing.
  • [COMMAND_EXECUTION]: Provides bash scripts and Python code that interface with the AWS CLI and SDK to perform management actions on cloud infrastructure.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 27, 2026, 05:53 AM