aws-cost-cleanup
Pass
Audited by Gen Agent Trust Hub on Apr 19, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes several Bash and Python scripts designed to execute commands via the AWS CLI and boto3 SDK. These scripts perform actions such as deleting EBS volumes, snapshots, and releasing Elastic IPs.
- [DATA_EXFILTRATION]: No network operations to non-whitelisted domains were found. The skill operates locally within the AWS environment to manage resources.
- [EXTERNAL_DOWNLOADS]: All external references point to official Amazon Web Services documentation (aws.amazon.com), which is a well-known and trusted service.
- [PROMPT_INJECTION]: The instructions do not contain attempts to override agent behavior, bypass safety filters, or extract system prompts.
- [DYNAMIC_EXECUTION]: The skill demonstrates standard script generation by creating local configuration files (e.g., lifecycle-policy.json), which is common for AWS automation tasks.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from AWS resource metadata (e.g., volume IDs and snapshot descriptions). While this represents a theoretical injection surface if resource tags are maliciously crafted by an external actor, the skill's workflow emphasizes human review in the 'Discovery Phase' before execution, mitigating this risk.
Audit Metadata