Tracks · Learn Claude
CLAUDE.md and Claude memory — what does Claude remember?
Claude has no memory between sessions except what you put back into the context — and the window is a budget you spend, not a bucket you fill. This track covers compaction versus truncation, CLAUDE.md and project instructions, long-running state, retrieval hygiene, and fresh-session handoff. Afterwards you can make Claude follow your SOP consistently instead of hoping.
What you learn in this track
The window as a budget, not a bucket
Every token you spend on context is a token you cannot spend on the answer.
Compaction vs. truncation
Truncation throws away the oldest turns; compaction rewrites them. Only one of those keeps the decisions you made.
CLAUDE.md and project instructions
A CLAUDE.md file is how you stop repeating your own standing instructions in every session.
Sessions & long-running state
State that matters must live outside the conversation, because the conversation will end.
Retrieval hygiene & citations
Retrieval fails quietly: the wrong chunk still produces a confident answer.
Making Claude follow your SOP
Getting Claude to follow your process consistently is a matter of placement and repetition, not politeness.
Fresh-session handoff
A good handoff note is written for a reader who has no memory of what you just did.
Guides in this track
- Claude memory: compaction or truncation when context runs out?
- What is CLAUDE.md and what should you put in it?
- CLAUDE.md, AGENTS.md or Cursor rules — which does what?
- Claude Projects, Claude memory — or a written handoff?
How the Socratic coach teaches this
Each module is a Socratic dialogue: the coach probes your reasoning with scenario questions, surfaces weak points, and generates flashcards and study notes from your own gaps. Mastery is tracked per concept, so you can see which ideas you actually handle and which ones still need work.
See where this track sits in the curriculum on the six tracks and why they are in this order, or open the ClaudeMaster.ai hub to start it.