AI coding

My multi-project review hub gives several local AI project offices one review front door, so I can focus on approval instead of ports.

My AI writing pipeline separates a stronger editor from a cheaper writer, so briefs and quality gates improve instead of prose rewrites.

AI debugging logs give my Cursor agent runtime evidence from browser, Electron, and Node apps through a local file it can read.

A session handoff system for Cursor: two small skills and a wiki folder that let a new agent session pick up where the last one stopped.

My Cursor workflow in five layers: rules, session memory, runtime logs, pipeline skills, and one review hub built on top of the editor.