The Agent Skills Directory

[SAFE]: The skill serves as a comprehensive guide for Ansible development best practices, emphasizing correct usage of the toolchain.
[CREDENTIALS_UNSAFE]: The skill explicitly instructs users to store secrets in Ansible Vault and prohibits hard-coding credentials or environment-specific values in playbooks and roles (references/common.md).
[COMMAND_EXECUTION]: While the skill discusses ansible.builtin.shell and ansible.builtin.command, it recommends using specific Ansible modules whenever possible. It also provides clear instructions on using creates, removes, and changed_when to ensure idempotency and correct error reporting when shell commands are unavoidable (references/developing-tasks.md).
[PRIVILEGE_ESCALATION]: The skill provides guidance on using become: true, which is a standard feature of Ansible for system administration tasks. It encourages explicit configuration of these settings for clarity and security (references/developing-playbooks.md).
[INDIRECT_PROMPT_INJECTION]: The skill processes user requests for Ansible code generation and review. While this is an ingestion point for untrusted data, the skill's instructions focus on generating secure, technical code and following established project standards, mitigating the risk of accidental malicious instruction execution.

developing-ansible