What this is
Claude starts every conversation from zero. That is not a bug, it is how the context window works, and it is why you keep re-pasting the same background before you can ask the real question. You pay for those tokens every single time.
This guide sets up a wrap up skill. At the end of a working session Claude writes a structured summary of what happened, and that summary goes into a NotebookLM notebook you keep forever. Next session, instead of re-pasting your whole history, Claude reads back only the part it needs.
One honest note before you start, because it changes how you set this up. NotebookLM has no public write API. Nothing can push a file into a notebook for you automatically. What the skill does is produce a clean, well shaped file in a folder you control. Getting that folder into NotebookLM is a one time setup, and after that it is either automatic through Google Drive or a five second upload. Anyone telling you otherwise has not tried it.
What you need
- Claude Code, on any plan
- A Google account with NotebookLM enabled
- A folder on your machine for the wrap ups. This guide uses ~/claude-memory
Free, for your email
Get the full setup steps + the copy-paste setup prompt
The rest of this guide (8 more sections) opens right here. The setup prompt, one paste into Claude that sets it all up, comes by email only. Use your real address.
Unlocks: Step 1. Make the memory folder · Step 2. Create the skill · Step 3. What the wrap up should actually capture · Step 4. Get the folder into NotebookLM · Step 5. Reading it back · +3 more
50,000+ follow where these guides come from. No spam, reply “unsubscribe” any time.