terraform-skill
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions are defensive and focus on security best practices for Infrastructure as Code. The workflow prioritizes diagnosis and minimal context loading, with mandatory validation steps.
- [REMOTE_CODE_EXECUTION]: Instructions include installation commands for well-known and trusted security and infrastructure tools (e.g., Trivy, TFLint, terraform-ls) from their official GitHub repositories. These operations are standard for the described development environment and target trusted vendors.
- [DATA_EXFILTRATION]: The skill explicitly identifies and mitigates secret management risks. It provides detailed guidance on preventing credential exposure in state files and logs, recommending the use of modern HCL features such as
write_onlyandephemeralvalues for state exclusion. - [INDIRECT_PROMPT_INJECTION]: The skill processes infrastructure code, state files, and CI/CD logs, which represents an ingestion surface for potentially untrusted data. The skill mitigates this risk by requiring strict validation workflows, including
terraform fmt -check,terraform validate, and manual plan review before any changes are applied.
Audit Metadata