The Agent Skills Directory

[SAFE]: The skill provides documentation and standard command-line/Python examples for model exportation using the official ultralytics package.
[EXTERNAL_DOWNLOADS]: The skill mentions that some backends may auto-install necessary toolchains (e.g., IMX, RKNN) upon first use. This is standard behavior for the ultralytics library to manage complex hardware-specific dependencies.
[COMMAND_EXECUTION]: Provides legitimate CLI examples for benchmarking (yolo benchmark) and validation (yolo val), which are standard operations within the Ultralytics ecosystem.
[INDIRECT_PROMPT_INJECTION]: While the skill processes model weights and data configurations, it does not ingest untrusted natural language data that could influence the agent's logic.

yolo-export