Notes on building software.
A personal blog of technical articles about software development.
Play a Sound When Claude Code Is Actually Done
Waiting on a long Claude Code task means watching the screen — tiring on the eyes. A Stop hook can play a sound when it finishes, but only if 'done' truly means every subagent and background job has ended. Here's how to build that honest signal.
A Claude Code Status Line That Tells You What Matters
On a long Claude Code session it's easy to run blind — context filling up, rate limits creeping toward a lockout, the wrong model or account active. Here's how to build a custom status line that surfaces all of it at a glance.
Running Two Claude Code Accounts on One Machine
I keep a personal and a work Claude Code account on one laptop and run both at once. Out of the box that isn't possible — here's the pain point, and the one environment variable that fixes it.
How I Manage Claude Code Memory: What to Commit, What to Keep Local
Claude Code reads instruction files before it does anything. Here is how I split them across user, project, and local files — which to commit, which to gitignore, and what kind of rule belongs in each.
The Explore, Plan, Implement, Verify Loop
A simple four-step loop keeps big tasks on track: explore first, plan, implement, then verify.
Editing a Plan Before You Run It
A plan is a draft you can shape. Refine it in plan mode until it is right, then approve.