The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted data from the Unity Editor (such as GameObject names and asset paths) which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Data enters the agent's context through the output of the editor-selection-get tool, which returns object names and paths from the hierarchy.
Boundary markers: The skill does not define explicit delimiters or instructions for the agent to ignore potentially malicious instructions embedded within the retrieved selection data.
Capability inventory: The agent utilizes the unity-mcp-cli tool, which provides capabilities to interact with and potentially modify the Unity Editor environment.
Sanitization: There is no documented sanitization or validation of the strings retrieved from the Unity Editor selection before they are processed.
[EXTERNAL_DOWNLOADS]: The skill documentation suggests installing a CLI tool to facilitate its operation.
Finding: The instructions recommend installing unity-mcp-cli globally using npm or running it via npx.
Context: This tool is a vendor-owned resource managed by the skill author to enable Unity integration.
[COMMAND_EXECUTION]: The skill's primary function involves the execution of shell commands.
Finding: The agent is instructed to use the unity-mcp-cli command to fetch data from the Unity environment.
Purpose: This execution is necessary for the skill to perform its intended task of retrieving editor selection state.

editor-selection-get