Content Retrieval Protocol

Token-efficient retrieval system using markdown-native section detection.

Summary

Core Principle: Never read a full file when partial content suffices.

Choose retrieval depth by need:

• Exists? → Grep (files_with_matches)
• Count? → Grep (count)
• Lookup? → Grep (-C:2)
• Overview? → Read (limit:40)
• Section? → Grep heading → Read (offset, limit)
• Full? → Read (justify first)

Quick Reference