terraform
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied infrastructure-as-code (HCL) files, which serves as an ingestion point for untrusted data. The capability inventory includes execution of CLI tools like terraform, tofu, tflint, and checkov. While the skill lacks explicit boundary markers for these inputs, it emphasizes rigorous validation through multiple scanning layers.
- [COMMAND_EXECUTION]: The skill uses various infrastructure management tools including terraform, OpenTofu, tflint, and checkov to format, validate, and plan changes. These commands are executed locally to generate artifacts for human review. The instructions explicitly restrict the agent from executing destructive commands or applying changes autonomously.
- [REMOTE_CODE_EXECUTION]: The skill facilitates the retrieval of external providers and modules from the Terraform Registry, GitHub repositories, and OCI registries. It mitigates associated risks by mandating exact version pinning in production environments and requiring the commitment of the dependency lock file (.terraform.lock.hcl) to verify the integrity of provider binaries.
Audit Metadata