aws-security
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFE
Full Analysis
- Read-only API Enforcement: The skill and its reference files strictly limit operations to non-mutating AWS CLI commands. It explicitly forbids the recommendation or invocation of any API that could modify resource state.
- Sensitive Data Handling: The instructions incorporate specific guidelines for handling sensitive information found in security findings, such as IP addresses and resource identifiers, by requiring summaries before raw data disclosure.
- Security Best Practices Integration: The skill provides educational context on logging (CloudTrail), encryption (KMS/TLS), and secure credential management, aligning with industry standards for AWS environment management.
- Data Posture Monitoring: The skill is designed to provide visibility into security posture across multiple AWS services (GuardDuty, Inspector, Macie, etc.) without introducing persistence or privilege escalation mechanisms.
- Indirect Prompt Injection Surface: The skill processes data from external security findings (e.g., threat intelligence, vulnerability descriptions). While this constitutes an ingestion surface for potentially untrusted data, the skill's focus on structured summarization and its lack of executive capabilities (read-only) mitigates the risk.
Audit Metadata