Lately I’ve been doing more AI agent assisted coding and vibecoding using Cursor and Claude Code. Oftentimes the agent will need some feedback about how a particular approach is working. This can involve me running the program, going through a particular UX flow, and then reporting the results back to the agent. Such manual verification is time consuming and kind of tedious when there are multiple rounds of back and forth.

A technique that helps is to prompt the agent to write a small helper script to test some functionality — like interfacing with an API or database or evaluating an algorithm with particular inputs. The agent can then invoke that small script itself quickly, view debug output, and plan its next move. It takes me out of that loop and makes the whole experience smoother. It’s not quite a unit test or full integration test, just some throwaway code to give the agent a feeler into the environment.