The Agent Skills Directory

Interaction with XR Environment Data: The skill processes data from the XR scene hierarchy and browser console logs to identify objects and verify interaction results. This represents a potential surface for indirect prompt injection where instructions could be embedded in the scene metadata or log messages. While this is a standard requirement for the skill's operation, it warrants consideration regarding the trust level of the environments being tested.
Ingestion points: The skill ingests data via the $ARGUMENTS variable, the mcp__iwsdk-dev-mcp__scene_get_hierarchy tool, and the mcp__iwsdk-dev-mcp__browser_get_console_logs tool.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the workflow for these data sources.
Capability inventory: The skill utilizes tools to manipulate XR device transforms (xr_set_transform), orient the view (xr_look_at), capture screenshots (browser_screenshot), and perform click actions (xr_select).
Sanitization: The skill does not implement specific sanitization or filtering for the data retrieved from the XR environment or logs before processing.

click-target