skills/terminalskills/skills/opentofu/Gen Agent Trust Hub

opentofu

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONPRIVILEGE_ESCALATION
Full Analysis
  • [PRIVILEGE_ESCALATION]: The instructions include the use of sudo for installing the OpenTofu package via snap, which is a standard administrative requirement for system-level package installation on Linux.
  • [COMMAND_EXECUTION]: The skill guides the agent through standard infrastructure management operations using the tofu CLI, including initialization, planning, applying configurations, and running tests.
  • [EXTERNAL_DOWNLOADS]: The skill references the official OpenTofu website and GitHub repository for installation resources. It emphasizes security by advising users to download and manually review installation scripts rather than piping them directly to a shell, and highlights the importance of verifying downloads with checksums or GnuPG.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes HCL configuration files and input variables, creating an inherent attack surface for indirect prompt injection common to infrastructure-as-code tools. The instructions mitigate the risk of data exposure by teaching client-side encryption for state and plan files, ensuring sensitive data processed by the tool remains protected.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 02:42 PM