Working with Claude
How I use Claude on my computer: memory, skills, advisor, hooks and the rules I set myself.
-
How I work with Claude Code
The setup behind most of what you read on this site: Claude Code on my Mac, one folder per project, and a few habits that keep it useful.
-
Memory: so Claude does not start from zero
Per-project notes that Claude writes and reads by itself, a global file for my personal setup, and a hook that keeps it all in sync between my computers.
-
Skills and plugins
Skills are packaged instructions Claude loads when a task calls for them. Which ones I use, which I do not, and what my usage report suggested.
-
The advisor: a second opinion
A stronger model that reads the whole conversation and reviews the plan before it is carried out, and again before the work is declared done.
-
Hooks: automatic actions
Hooks are commands the app runs by itself on certain events. Mine keep my memory in sync, guard the servers and the secrets, check Python and block a bad push.
-
The rules I set myself
Claude is fast and capable, so the guard rails matter more, not less: when it may act on its own, when it must ask, and what never goes through the chat.