The Agent Skills Directory

[SAFE]: The skill serves as a technical guide for the MuJoCo Studio framework, providing legitimate instructions for simulation and visualization development.
[DATA_EXFILTRATION]: The document describes the launch_web functionality, which starts a local HTTP/WebSocket server (typically on port 8080) to stream UI and simulation state to a browser-based WebViewer. This is a core, documented feature of the framework intended for remote visualization and does not represent unauthorized data exfiltration.
[INDIRECT_PROMPT_INJECTION]: The skill describes an architecture that processes simulation models (XML MJCF files) and external assets. While these ingestion points represent a theoretical surface for indirect prompt injection if an agent were to process untrusted models, the skill itself defines a standard developer toolset and does not contain malicious instructions.

mujoco-studio