The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions direct users to clone and synchronize code from an external repository at https://github.com/sibilleb/AAP-Enterprise-MCP-Server. This repository is not associated with a recognized trusted organization or well-known service.
[COMMAND_EXECUTION]: The skill requires the execution of python3 and ansible-lint via a wrapper script (ansible-lint.py) using uv run. This creates a dependency on the integrity of the external wrapper script and the local environment.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection by processing external Ansible playbooks, roles, and project files.
Ingestion points: Tools such as lint_playbook, lint_file, lint_role, and analyze_project ingest user-supplied Ansible content.
Boundary markers: The skill documentation does not specify the use of delimiters or instructions to ignore embedded commands within the playbooks being analyzed.
Capability inventory: The skill executes subprocesses via ansible-lint to validate the ingested content.
Sanitization: There is no evidence of sanitization or filtering of the content before it is processed by the linting engine.

aap-lint