The Agent Skills Directory

[COMMAND_EXECUTION] (HIGH): The skill provides patterns for executing ansible-playbook with become: true, which allows for privileged command execution (root access) on target systems. While standard for Ansible, this capability represents a significant security risk if the input is untrusted.\n- [PROMPT_INJECTION] (HIGH): The skill is susceptible to Indirect Prompt Injection. It processes external Ansible playbook files (.yml) and configuration files (.ansible-lint) which could contain malicious instructions designed to subvert the agent's logic or exploit its execution environment.\n
Ingestion points: Playbook files and lint configurations stored in the ansible/ directory.\n
Boundary markers: Absent; the skill does not use delimiters or instructions to prevent the agent from obeying commands embedded within the data files.\n
Capability inventory: Full system command execution via uv run ansible-playbook and ansible-lint, including privileged operations on targets.\n
Sanitization: No sanitization or validation of the YAML content is specified before execution or linting.\n- [EXTERNAL_DOWNLOADS] (MEDIUM): The GitHub Actions documentation includes an unversioned installation of the uv tool via pip install uv. This practice can expose the environment to supply chain attacks if the package registry is compromised or if malicious versions are published.

ansible-testing