Yahya Rogers

A few months back I set up custom instructions for Claude stop opening with agreement, tag its confidence on claims, push back when I'm wrong instead of validating me. Then I mostly forgot about it. Kept getting the normal helpful-assistant experience.

Last weekend, mid-project, it started actually doing what I'd asked. Pulled up my own past chats and launch checklist to see what I'd actually committed to. Called out that I'd blown past my own launch deadline without noticing. Refused to let me reprioritize a payment-provider decision over just talking to five actual users.
Told me a "quick bug fix" I described was actually five missing subsystems.

Was it mean? Maybe, some of it. But all of it was annoying in the specific way that's actually useful the way a good cofounder or an honest advisor is annoying.

Felt strange, Do I need this? Should I tone down the instructions?
I don't know if everyone can replicate this exactly.
Curious who else has tried shaping how their AI pushes back, not just what it outputs.

Changes:

- Added ability to select files with box selection

- New setting "Enable box selection" which enables/disables this feature

- New setting "Increase file view gaps" which increase the empty space between file cards so it becomes easier to start box selection

Sometimes you set out to solve a problem and end up solving the wrong one.

I got feedback that the one-click Cloudflare deploy had friction. Cloudflare won't let you set the Vectorize dimensions and metric through wrangler.json, so I went looking for a workaround. One hack led to another. The solution I landed on basically amounted to "hope this works, and if it doesn't, install wrangler and run a command manually."