You run it. If it gives an error or the output isn't what you want, you tell Chatgpt to try again.
Also using 4 and not 3.5.
I imagine there is so much pointless discussion online with one person talking about 3.5 and the other person talking about 4.
I think a good way to verify that the commands are actually doing something instead of the LLM just generating non-existent parameters is to provide a dry run mode. The actual correctness of the generated command - granted it's valid - performing as it should however, is a whole different issue.
I also have had much success with FFMPEG and other bash commands - I just ask it to explain what the command does step by step as I want to learn a bit as well.