Collection · 5 videos · 5 min
Claude: Worth Knowing
Short tips on using Claude well. Each one shows a single control in the app, says what it does and what it costs to ignore, and ends on the move to make. The screens are generic copies of the real pages, so nothing in them belongs to anyone.
Every tip stands on its own and runs under a minute, so start with the one you've wondered about. Independent explainers, not affiliated with Anthropic.
-
0:56 Decide the model and effort before you ask
The model menu holds two choices, the model and the effort level, and Claude pays for both on every reply after you make them. What a late change costs, and why to set them once, at the start.
-
1:00 Decide whether your chats train Claude
One switch in Settings, under Privacy, decides whether your chats and coding sessions train Anthropic's models, and how long they are kept. What it covers, what gets round it, and why to decide it before the next chat.
-
1:08 Edit what Claude remembers about you
Claude keeps a memory of you, written from your chats and read at the start of every new one. It is listed by topic in Settings, under Memory, and you can edit or delete any of it. What a stale entry costs, the one switch worth knowing, and the move.
-
1:05 Make it an artifact when you will use it again
A reply in a chat scrolls away. An artifact opens beside the chat, can be edited in place, and is saved in the Artifacts tab where you can find it from any conversation. What an artifact is, what a plain reply costs you, and when to ask for one.
-
1:03 Decide who can open an artifact before you share it
Every artifact starts private. The Share control sets who can open it, only you, named people, or anyone with the link, and on a work plan a shared chat artifact can carry the chat's files with it. What each level means, what the wrong one costs, and how to take a link back.