The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the unity-mcp-cli command-line tool to interface with the Unity Editor's SceneManager. This is the primary intended function of the skill.
[EXTERNAL_DOWNLOADS]: The skill documentation refers to the unity-mcp-cli package available via the NPM registry, which is a required dependency for executing the skill's commands.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection where untrusted data could influence command execution.
Ingestion points: The name parameter provided to the tool is ingested directly into a shell command string.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious content within the scene name.
Capability inventory: The skill possesses command execution capabilities through the unity-mcp-cli wrapper.
Sanitization: The skill definition does not specify sanitization or validation logic for the name input before it is interpolated into the CLI command.

scene-unload