aws-cost-optimizer

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [Command Execution] (SAFE): The skill includes numerous AWS CLI commands (e.g., aws ec2 describe-instances, aws s3 ls) intended for resource discovery and management. These commands are typical for cloud administration and align with the skill's stated purpose.
  • [Data Exposure & Exfiltration] (SAFE): No sensitive data access, hardcoded credentials, or external network exfiltration patterns were identified. All commands target standard AWS API endpoints using user-provided credentials.
  • [Indirect Prompt Injection] (LOW): The skill possesses a vulnerability surface for indirect prompt injection by reading resource metadata (like tags and names) through aws ec2 describe-instances and aws rds describe-db-snapshots.
  • Ingestion points: Resource tags and metadata fetched via aws CLI commands in SKILL.md.
  • Boundary markers: Absent; there are no delimiters or instructions to ignore embedded content in resource tags.
  • Capability inventory: Destructive actions including terminate-instances, delete-volume, and delete-snapshot.
  • Sanitization: Absent; the skill does not specify validation or sanitization for data retrieved from the AWS environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:24 PM