
This is the skill you want when you're actively shaping what things mean in your codebase, not just reading about them. It maintains a CONTEXT.md glossary and architecture decision records, challenging fuzzy language the moment it appears and forcing you to be precise about domain boundaries. The discipline here is aggressive: it stress-tests terms with edge cases, catches contradictions between code and conversation, and updates the glossary inline rather than batching changes. It only creates ADRs when decisions are hard to reverse, genuinely surprising, and involve real trade-offs. If you've ever had three people in a room using "account" to mean different things, this catches that immediately.
npx -y skills add mattpocock/skills --skill domain-modeling --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.