Collecting is not the finish line
Most people do not fail to save things. They save too widely: a chat log, a bookmark folder, a download directory, and one conclusion nobody can find again.
LLM-notebook does one job — it puts four kinds of material in one place, and makes each of them askable the moment it lands.
Four sources, four entry points
- LLM chats: hit "Save chat" on ChatGPT, Claude, DeepSeek and friends. The whole thread is filed with its platform label attached. No copy-paste.
- Web links: paste a URL. Title, body and images are fetched and parsed into a searchable item.
- Files: PDF, Word, PPT, Markdown, images. Drop them in and they are parsed and filed.
- Agent sessions: with the skill installed, tell your agent "save this to LLM-notebook" and the session becomes a note.
Four doors, one room. After that you stop caring which app it came from.
Filing is the hard part
Saving is one click. Filing decides whether you can still find it three months later. So parsing extracts title, source, time and body structure up front, and every item carries those dimensions from birth.
Found is what makes it collected. Unfindable is just a copy.
Three things worth doing
- Add one line of your own when you save it. You will thank yourself in three months.
- Layer with folders, not tags. Keep tags for cross-topic keywords.
- Ask a question right after saving — that turns a stored item into a used one.
Install the browser extension or the skill and all of this happens inside the workflow you already have.
Keep reading
CollectLLM-notebook: turning flowing LLM conversations into knowledge assets
A sharp answer in ChatGPT, a workable engineering practice in Gemini, reusable reference material in Qwen — and a week later it is scattered across four chat histories, impossible to search and impossible to edit. LLM-notebook collects LLM conversations into one library and settles them into notes you can read, export and share.
Read post
AskLLM-notebook: turn a YouTube video into material you can question
YouTube has whole courses on it, but video is linear: to find the sentence the instructor said ten minutes ago you drag the scrubber and hope, and there is nowhere to ask about the part you did not follow. LLM-notebook turns a YouTube link into material you can question — the video plays in the workspace, the subtitles are laid out by paragraph, and anything you do not understand you select and ask about. Answers cite the exact passage, and one click jumps back to that moment.
Read post
ChangelogChangelog
From the first version on main to today, logged week by week. Every entry comes from a real commit — features, but also the mistakes and the rollbacks.
Read post