This file is a working-state handoff so development can continue on another
machine (which has this repo but NOT the developer’s ~). It captures the
architecture, what works, what is untested and the design decisions made so far.
This file is to be maintained by Claude. When a release is being made, claude should be prompted to update this document.
Companion docs: AI-AGENTS.org (agent guidance / layering), README.org (overview).