The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard shell commands for the installation of Ansible and its dependencies via system package managers (apt, yum) and pip. It also utilizes the Bash tool to execute ansible, ansible-playbook, and ansible-galaxy commands, which is consistent with the skill's primary function as an automation expert.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process and execute configuration management scripts (YAML playbooks) and dynamic inventory scripts (Python). This creates an attack surface where untrusted configuration data could influence agent behavior.
Ingestion points: Processing of user-provided Ansible playbooks (SKILL.md), inventory files, and Jinja2 templates.
Boundary markers: No specific delimiters or warnings for embedded instructions are provided in the documentation examples.
Capability inventory: Access to the Bash tool for running Ansible commands, Write for creating playbooks, and network operations through modules like get_url and uri (references/PLAYBOOKS.md).
Sanitization: The skill encourages best practices such as no_log: yes to prevent data exposure but does not specify input validation for user-provided automation scripts.
[SAFE]: The skill implements best practices for secret management by demonstrating the use of Ansible Vault and environment variable lookups (lookup('env', '...')) rather than hardcoding sensitive credentials. All external package references are for well-known and trusted DevOps tools.

ansible-expert