terraform-state-operations
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- COMMAND_EXECUTION (LOW): The skill generates shell commands for infrastructure management using the Terraform CLI and system utilities like
cpanddate. While these commands are powerful, they are restricted to the skill's primary purpose and include defensive patterns such as timestamped backups and verification steps. - DATA_EXPOSURE (LOW): The skill interacts with
terraform.tfstatefiles. These files are sensitive as they contain infrastructure metadata; however, access is required for state management, and the skill does not demonstrate patterns of sending this data to external or untrusted domains. - INDIRECT_PROMPT_INJECTION (LOW):
- Ingestion points: User-provided resource names, module paths, and IDs are used to construct CLI commands in
SKILL.md. - Boundary markers: Present. The 'Approval Workflow' and 'Verification Checklist' provide clear human-in-the-loop checkpoints before any command execution.
- Capability inventory: Local file operations (
cp, redirection) and remote state synchronization (terraform state pull/push) across all command examples inSKILL.md. - Sanitization: The skill relies on 'Verification after [operation]' steps and user review rather than automated sanitization to prevent command injection via resource names.
Audit Metadata