aws-cloudformation
Pass
Audited by Gen Agent Trust Hub on Jul 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements strong security boundaries by explicitly instructing the agent to treat all user-provided template content (including metadata and comments) as untrusted data, effectively mitigating prompt injection risks.
- [EXTERNAL_DOWNLOADS]: The skill references and utilizes well-known, official AWS tools including
cfn-lintandcfn-guard. Documentation lookups target the official AWS documentation domain (docs.aws.amazon.com). All external references are to trusted sources. - [COMMAND_EXECUTION]: Shell commands are used to invoke standard development tools (
awsCLI,cfn-lint,cfn-guard). All such operations are transparent and include mandatory user approval steps before execution or installation of missing dependencies. - [DATA_EXFILTRATION]: No patterns of data exfiltration were detected. The skill interacts with the AWS environment using the user's local credentials and does not transmit data to third-party domains.
- [CREDENTIALS_UNSAFE]: The skill promotes secure credential management, explicitly instructing the user to avoid hardcoded secrets in templates and recommending the use of AWS Secrets Manager and SSM Parameter Store dynamic references.
Audit Metadata