When you talk to your AI colleague, it should understand your shortcuts and nicknames. Memory Management gives your AI a two-tier memory system. It decodes shorthand, acronyms, and internal language so your AI gets you like a real coworker.
The first tier is a hot cache called CLAUDE.md. It holds the top people, terms, and projects you use daily. The second tier is a memory directory with full profiles and a complete glossary. This keeps the hot cache lean while supporting unlimited growth.
Your AI checks the hot cache first, then searches the deep memory. If something is still unknown, it asks you and remembers the answer. This turns confusing requests into clear actions.
Global
mkdir -p ~/.claude/skills/memory-managementProject
mkdir -p .claude/skills/memory-managementSource Repository
Obsidian Baseskepano/obsidian-skills
Build powerful database views inside Obsidian with filters formulas and summaries
Compressjuliusbrussee/caveman
Shrink memory files into caveman speak to save tokens and preserve code
Dbs Restoredontbesilent2025/dbskill
Continue your diagnosis where you last left off without repeating yourself
Remembergithub/awesome-copilot
Save lessons learned and build a smart knowledge base that grows with you
Remembering Conversationsobra/episodic-memory
Search past conversations to recall decisions, solutions, and avoid repeating mistakes
Rememberrohitg00/agentmemory
Save important insights with tags and recall them later easily
Session Historyrohitg00/agentmemory
See your recent session history in a clean timeline instantly
Recallrohitg00/agentmemory
Quickly recall past sessions and learnings from your agent's memory