The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill uses the execute_tool_script MCP tool to run Python code directly within the Unreal Engine environment. This pattern is employed in actors.py for batch rendering, camera.py for retrieving engine state, and panels.py for spawning text actors.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data including network topology definitions, interface inventories, and PagerDuty incidents.
Ingestion points: Data is ingested through TopologyGraph objects in renderer.py and incident lists in incidents.py.
Boundary markers: None detected; data is processed as raw strings for visualization.
Capability inventory: The skill can spawn/destroy actors, modify materials, and execute scripts in the engine.
Sanitization: Input data is JSON-serialized but not otherwise validated for malicious instructions before being rendered as 3D text labels.
[COMMAND_EXECUTION]: The skill interacts with the local environment by making HTTP requests to a UE5 MCP server to manage the 3D scene and its components.

ue5-network-viz