[BUG] Claude is useless for testing anything

Preflight Checklist

• [x] I have searched existing issues and this hasn't been reported yet
• [x] This is a single bug report (please file separate reports for different bugs)
• [x] I am using the latest version of Claude Code

What's Wrong?

Any feature Claude makes - it always refuses to test it. Despite env being clearly test environment, being sandboxed, Claude made this feature.
Yet he refuses to login into test account to test the feature he just made!!!!!!

What Should Happen?

Tool must behave as tool on request, not refusing because he has bad mood.
Every single other AI provider works perfectly, when testing the features. Yet Claude is a special boy.

Why all other AI providers are perfectly fine testing everything, yet Claude is dumbed down? Why are we paying for cut down tool?
I dropped Cursor IDE to try Claude, and it's worse at all places forcing me to regret the decision.

Error Messages/Logs

Steps to Reproduce

ask Claude to login into test account in test environment - Claude says "nope, i won't login, I can't, I'm forbidden, I'm now allowed"

Claude Model

None

Is this a regression?

I don't know

Last Working Version

_No response_

Claude Code Version

2.1.293 (desktop app)

Platform

Anthropic API

Operating System

macOS

Terminal/Shell

Terminal.app (macOS)

Additional Information

_No response_