aws-security-audit
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external AWS APIs which can contain attacker-controlled content in metadata or resource names.\n
- Ingestion points: AWS CLI command outputs (IAM reports, EC2 descriptions, S3 bucket metadata) and boto3 API responses.\n
- Boundary markers: No specific boundary markers or instructions are used to separate external data from the agent's instructions.\n
- Capability inventory: Extensive use of shell commands (aws CLI) to query environment state.\n
- Sanitization: No sanitization of resource names or metadata is performed before displaying them to the agent.
Audit Metadata