Building developer tools for Cursor

Cursor Prompt Enhancer v2 and YAP Cursor Extension came from the same observation: developers often know what they want, but their first prompt is under-specified. The goal was not to make prompting feel magical. It was to reduce the number of times a developer has to restate constraints that should have been included from the start.

The Open VSX launch gave me a useful feedback loop. Downloads are a rough signal, but extension behavior is easier to judge by whether it fits into existing muscle memory. If a tool asks for too much ceremony, developers stop using it even if the underlying idea is good.

The biggest design choice was keeping the extensions close to the editor workflow. Prompt improvement should feel like sharpening the instruction you were already about to send, not like opening a separate product. That meant the scope had to stay tight and the output had to be immediately editable.

Reaching 1K+ total downloads was a good validation, but the more important result was a better mental model for developer automation: the tool should raise the floor of a request without flattening the developer's intent.