A knowledge base for architecture that can be checked.

Essays on agents, declared intent, and the difference between reading a codebase and knowing what it is allowed to do. Scan by date.

A million-line source tree versus a change-scoped architecture graph.

An agent that reads faster still can't find what was never written down.

Reading code does not scale. Neither does delegating it.

Read the post →
A CLAUDE.md file of architectural prose versus an architecture.calm.json graph.

Get the Architecture Out of Your .md Files

AI agents need an architecture model, not more markdown.

Read the post →