Senior Developer Explanation
I'll explain this code as a senior developer would, focusing on the why behind decisions.
Token Optimization
Expected range: 350–2,500 tokens (initial), 100 tokens (cache hit)
Caching: Caches architecture analysis in .claude/cache/understand/architecture.json for 7 days.
Early exit: Returns immediately if the target is too simple to require senior-level explanation.
Patterns used: Grep-before-Read, early exit, caching, progressive disclosure