aws-s3-eks
Pass
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes aws, kubectl, and eksctl commands via bundled bash scripts to manage cloud resources. It dynamically generates IAM policy documents using variable interpolation and heredocs before applying them via the AWS CLI.
- [PROMPT_INJECTION]: An indirect prompt injection surface exists where user-provided arguments are incorporated into shell commands and IAM policy definitions. Ingestion points: Parameter parsing in scripts/create-s3-bucket.sh and scripts/setup-pod-identity.sh. Boundary markers: None identified. Capability inventory: Access to system-level CLI tools including aws, kubectl, and eksctl. Sanitization: Variables are quoted within the bash scripts to prevent simple word splitting, but the skill lacks validation of inputs against malicious characters or cloud provider naming constraints.
Audit Metadata