Two minor gripes with the Gemini CLI tool. I know, I know … it just came out.

1. There’s no init command to automatically get it to generate its context file. I asked it to do it, and it created an empty file, and then I had to follow up with a second prompt to get it to update the file with helpful information. I’m never going to use that until it’s easier.

2. I ran a few prompts, and it automatically switched to the flash model, and there’s no way to control what model it uses. Like maybe I’m okay with a bit slower responses if it means I can use the pro-model. I’d like the ability to be able to control this, and there doesn’t seem to be a way to switch between the models.