ansible

Pass

Audited by Gen Agent Trust Hub on May 26, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing Ansible from official package managers (pip, apt, dnf, homebrew) and downloading roles/collections from Ansible Galaxy (e.g., geerlingguy.java, amazon.aws). These are standard practices for the tool.
  • [COMMAND_EXECUTION]: The skill provides numerous examples of shell commands for infrastructure automation, including ansible-playbook and ad-hoc commands. It includes safety guardrails recommending dry-runs (--check) and targeting subsets (--limit).
  • [CREDENTIALS_UNSAFE]: The documentation references sensitive file paths such as SSH keys (/.ssh/) and vault password files (/.vault_pass) for configuration purposes. Importantly, it emphasizes using ansible-vault to protect secrets rather than storing them in plaintext.
  • [PROMPT_INJECTION]: As a configuration management tool, the skill is designed to process external data. Ingestion points: inventory files and variable directories (e.g., inventory.ini, group_vars/ mentioned in references/inventory-and-variables.md). Boundary markers: None. Capability inventory: Remote command execution and file management (documented in references/command-cookbook.md). Sanitization: None. While this creates a surface for indirect prompt injection, it is the primary intended function of the tool and is addressed by safety instructions regarding privilege escalation.
Audit Metadata
Risk Level
SAFE
Analyzed
May 26, 2026, 05:02 PM
Security Audit — agent-trust-hub — ansible