The Agent Skills Directory

[SECURITY_BEST_PRACTICES] (LOW): The skill recommends setting host_key_checking = False in the ansible.cfg file. This disables SSH host key verification, making the system vulnerable to Man-in-the-Middle (MitM) attacks during automated connections.
[INDIRECT_PROMPT_INJECTION] (LOW): The skill demonstrates a pattern where untrusted data from environment analysis is interpolated into configuration files and commands without sanitization.
Ingestion points: User-provided server IP addresses, hostnames, and user credentials gathered in Phase 1.
Boundary markers: Absent; the templates do not use delimiters or warnings to ignore instructions embedded within these variables.
Capability inventory: The skill uses ansible-playbook and shell commands, which possess the capability to execute tasks with elevated privileges on remote servers.
Sanitization: Absent; the skill does not suggest any validation or filtering of the gathered data before it is used in the inventory or ansible.cfg files.
[COMMAND_EXECUTION] (SAFE): The skill instructs the agent to run standard Ansible commands and shell utilities for project management. This is the primary intended purpose of the skill.
[DATA_EXPOSURE] (SAFE): The inventory example references the sensitive path ~/.ssh/id_rsa. This is appropriate for Ansible authentication and does not constitute unauthorized exposure in this context.

ansible-interactive