aws-infra

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides comprehensive instructions for managing AWS infrastructure using standard tooling including the AWS CLI and the boto3 Python library.
  • The provided snippets allow the agent to perform administrative tasks such as listing resources, deploying CloudFormation stacks, managing EKS clusters, and invoking Lambda functions.
  • While these are high-privilege operations, they are explicitly aligned with the skill's stated purpose of AWS infrastructure management.
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface: The skill interacts with external data (AWS resource metadata, stack events, and logs) which constitutes an ingestion point for potentially untrusted data.
  • Ingestion points: list_resources, describe_stacks, describe_stack_events, list_functions in SKILL.md.
  • Boundary markers: No explicit boundary markers or 'ignore' instructions are used when processing resource data.
  • Capability inventory: The skill has the capability to create resources (create_resource), invoke code (invoke), and update local configuration (update-kubeconfig).
  • Sanitization: No input sanitization or validation of the ingested AWS metadata is present in the provided templates.
  • Note: This represents a standard attack surface for infrastructure management tools rather than a malicious implementation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 03:29 AM
Security Audit — agent-trust-hub — aws-infra