The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to run the TAO Toolkit container via the auto_label CLI. It also provides instructions for launching a vLLM server using Docker or pip for local inference.
[EXTERNAL_DOWNLOADS]: The skill references the download of the vllm package from public registries and the vllm/vllm-openai Docker image. These are well-known resources in the AI/ML community.
[SAFE]: All identified command executions and external downloads are necessary for the skill's stated purpose of image processing and VLM inference. The skill uses standard practices for secret management by requesting API keys through environment variables.

tao-generate-referring-expressions