oma-tf-infra
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill follows established industry security standards for Infrastructure-as-Code, including specific guardrails to prevent hardcoding secrets and mandates for manual review of destructive infrastructure changes.
- [COMMAND_EXECUTION]: The skill employs the Terraform CLI and security tools like Checkov and Trivy. These operations are essential for infrastructure management and are performed within a controlled command path defined in the skill instructions.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a potential attack surface through the ingestion of external data such as existing Terraform files and plan outputs.
- Ingestion points: The agent reads project-specific configuration files (.tf, .tfvars) and Terraform plan outputs during its preparation and acquisition phases.
- Boundary markers: The skill instructions emphasize identifying module boundaries and risk levels, though it lacks explicit technical delimiters for all ingested data.
- Capability inventory: The skill is capable of writing HCL code, executing shell commands for infrastructure validation and scanning, and communicating with cloud provider APIs for remote state management.
- Sanitization: Infrastructure code is validated using standard CLI tools (terraform validate and fmt), and any significant resource modifications require explicit user authorization.
Audit Metadata