eksctl
Warn
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: MEDIUMCOMMAND_EXECUTION
Full Analysis
- [Command Execution] (MEDIUM): The skill is centered on executing eksctl commands that modify cloud infrastructure. These commands can create, delete, or modify clusters and nodegroups. While this is the intended purpose, it involves high-impact command execution.
- [Privilege Escalation] (MEDIUM): Includes operations like 'eksctl create iamserviceaccount' and 'eksctl create accessentry'. These can be used to grant significant permissions within the AWS environment or the EKS cluster, which are high-privilege actions.
- [Indirect Prompt Injection] (LOW): The command templates utilize unquoted placeholders (e.g., , , ) that ingest untrusted data. This creates a surface for command injection if the calling agent does not sanitize user inputs. 1. Ingestion points: Placeholder variables in bash blocks within SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Shell command execution via eksctl and aws CLI for cluster and IAM management. 4. Sanitization: No sanitization or validation logic is defined in the skill.
Audit Metadata