The Agent Skills Directory

[SAFE]: The skill acts as a central hub or router, delegating user inputs to other internal skills based on keywords (e.g., routing 'inbox' to '/ops-inbox'). This is a standard architectural pattern for complex agent skills.
[SAFE]: The skill reads local configuration files (preferences.json and daemon-health.json) from the plugin's data directory. This is expected behavior for an operational dashboard to determine its state and available features.
[SAFE]: It utilizes an experimental 'Agent Teams' feature to parallelize tasks using sub-agents (TeamCreate, Agent). The instructions provide a clear template for coordinating these agents to gather status updates from different business areas.
[SAFE]: The skill does not perform network operations, hardcode credentials, or execute arbitrary remote code. All tool usage is focused on internal routing and reading its own configuration.

ops