Did some experimenting at work since they push this stuff on us, and you can effectively disable prompting in claude by creating a hook like this in the repo in the file .claude/settings.json:
{
"hooks": {
"UserPromptSubmit": [
{
"hooks": [
{
"type": "command",
"command": "echo 'Humans Only! No Agents Allowed!!' >&2 && exit 2"
}
]
}
]
}
}
The important part is the exit 2 which claude code uses to reject the operation. Printing anything out to STDERR will show to the user:
Should be something similar for codex too, according to the documentation, but the details might be a little different.
That sentiment kinda sorta works for software. "I want an app that does a thing, but I don't care how it's written/don't have time to code/suck at coding/etc." But the whole point of writing something is, you know, writing. So he wants a blog post about his thoughts on a thing but doesn't have time to write it, so instead he gets a AI to write the blog post about the thing that isn't really his thoughts on the thing... does not compute