My local AI assistant kept asking permission for every tiny step, so I added conversation-level auto-approve without removing the real safety boundaries.
I like local AI assistants that ask before they do risky things.
I do not like approving the same task six times in a row.
That was the failure mode I kept hitting while working on CliGate. A normal task would start with one reasonable command, then another, then another, and every tiny step wanted a fresh confirmation.
So I added a blunt but useful switch:
/yolo
A local AI assistant needs task-scoped approval memory, or one real task turns into a pile of repeated permission prompts.
A forged confirmation, falsely green tests, and the acceptance gate I built afterward.
With these prompts, my AI tools get away from being a Yes-Man
Two ways to add human approval to AI agents in Genkit - discretionary interrupts vs restartable tools with deterministic gates -…
Why I split my local AI assistant into experience, control, execution, observation, and memory planes instead of one giant loop.
This isn't about AI being dangerous. It's about a habit most of us have developed without...
