The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the local Unity environment by retrieving the name of the current undo group via the history_get_current tool. While this data could theoretically contain malicious instructions (e.g., if a previous operation was named to trigger an injection), the skill's capabilities are restricted to native Unity undo/redo operations, and there is no evidence of the agent executing content from these names.
Ingestion points: The history_get_current skill returns currentGroup (the name of the undo group) from the Unity Editor environment (SKILL.md).
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present when processing the retrieved undo group name.
Capability inventory: The skill is limited to native undo and redo calls within the Unity Editor; it does not perform network operations, file system writes, or arbitrary command execution.
Sanitization: No evidence of sanitization or validation for the currentGroup name returned by the Unity API.

unity-history