Use the AI Brain to remember

The AI Brain is the part of Tase that holds things you want the assistant to know permanently. It is split into five sections, each with a different job, so you can find a fact later without digging through chat history.

The five sections and what they are for

• Notes - quick jots and reference text you want to keep.
• Ideas - half-formed thoughts you might act on later.
• Knowledge - structured facts, how-tos, and things you looked up once and want to avoid looking up again.
• Prompts - reusable prompts you want to run again without retyping.
• Memories - facts Tase extracts automatically from your conversations, tagged by category and importance.

Walkthrough: capture and recall

1. 1

   Open the AI Brain sidebar

   On iPhone, open the sidebar and pick Notes. On web, navigate to the Brain view. You will see the five sections listed.
2. 2

   Create a note

   Tap add and type something you want to keep, for example "Apartment building code: 4471." Tap save. It is now searchable and permanent.
3. 3

   Promote a note to Knowledge

   If the note is a durable fact rather than a quick jot, move it to Knowledge. This keeps your Notes list short and your reference material in one place.
4. 4

   Have a normal conversation

   Go back to the main chat and mention something personal in passing, like "I am allergic to penicillin." You do not need to do anything special.
5. 5

   Watch the memory appear

   Within a few minutes, open the Brain and check Memories. Tase extracts the fact, writes it in third person ("User is allergic to penicillin"), tags it, and ranks it by importance.
6. 6

   Search across everything

   Use the Brain search to find anything across all five sections. This is faster than scrolling chat history when you need a specific fact.

How memory extraction decides what to keep

Tase does not save every sentence you say. The extraction service evaluates each conversation, picks out durable personal facts, and assigns an importance level from 1 to 3. High-importance memories (allergies, family, recurring commitments) carry more weight in future conversations than low-importance ones (a one-off preference).

The result is an assistant that gets easier to work with over time. The more you use it, the more context it quietly accumulates, and the less you have to repeat yourself.