aws-cost-optimize

Pass

Audited by Gen Agent Trust Hub on Aug 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches cost management guidelines and best practices from official AWS documentation at docs.aws.amazon.com.- [COMMAND_EXECUTION]: The skill executes multiple AWS CLI commands to discover resources (EC2, RDS, Lambda, S3, etc.) and retrieve utilization metrics from CloudWatch and Cost Explorer.- [DATA_EXFILTRATION]: The skill reads infrastructure metadata, usage patterns, and monthly cost data from the AWS environment and local Infrastructure-as-Code files, which is then summarized and posted as issues to a target GitHub repository.- [PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests and processes data from local configuration files and AWS API responses to generate its recommendations.
  • Ingestion points: Local IaC files (.tf, .yaml, .json, .ts) and AWS CLI command outputs.
  • Boundary markers: None explicitly defined in the instructions.
  • Capability inventory: Local file reading, AWS CLI execution, network fetching via fetch, and GitHub issue creation via the GitHub MCP.
  • Sanitization: No specific validation or sanitization of the infrastructure data is performed before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 7, 2026, 11:25 AM
Security Audit — agent-trust-hub — aws-cost-optimize