aws-cost-optimization
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill makes extensive use of the AWS CLI (
aws ec2,aws ce,aws cloudwatch,aws compute-optimizer) to query and manage cloud infrastructure. While these are legitimate tools for the stated purpose, they represent a significant capability for resource deletion and modification.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from the AWS environment (e.g., resource tags, instance names, and usage metrics). There is a theoretical surface for indirect prompt injection if an attacker-controlled AWS resource contains malicious instructions that the agent might inadvertently execute or adopt.\n - Ingestion points: Data returned from commands like
aws ec2 describe-instancesandaws compute-optimizer get-ec2-instance-recommendations(found inSKILL.md).\n - Boundary markers: Absent from the prompt templates or instructions.\n
- Capability inventory: Shell command execution via the terminal and file write/copy operations (
cp,date,awsCLI) as seen inSKILL.md.\n - Sanitization: No explicit sanitization or validation of the AWS API output is performed before it is used in report generation or further commands.
Audit Metadata