One situation I found is "AI: it works on my browser" - "User: not on mine" -> and "AI: ok let me connect to your browser so I can run some test".
It's scary but effective - the alternative is a continuous copy&paste of logs from the dev console
[delayed]
Are you burning tokens having your agent reinvent the wheel every time it needs a wheel?
Shell execute and apply patch are the only two tools you need if your reasoning model is powerful enough.