# Brain: Persona & Knowledge

The **Brain** page (avatar menu → Brain) is where you teach the twin about you. Two panels:

## Persona — *how* I write

Your voice profile: tone, brevity, word choice, sign-offs (or lack of them).

By default Twico **learns this from your last 90 days of outgoing Slack messages** the first time you connect a workspace. You'll see the inferred persona in plain English — short sentences capturing the rules: "lowercase first word", "no greetings", "uses 'lol' but not exclamation marks", "closes longer messages with a dash + initial".

You can edit it freely. Anything you write here flows into the system prompt for every reply.

## Knowledge — *what* I know

A list of durable facts the twin keeps in mind. There are three sources:

1. **Auto-memory** — after every successful reply the agent decides whether the exchange contained anything worth remembering ("Sarah prefers Slack DMs over email", "Friday standup is at 9am PT") and saves it here automatically. Per-channel 30-minute cooldown so a busy thread doesn't flood the list.
2. **Manual notes** — type a title + body and click Save. Use this for things the twin should always know: pricing, project names, your role.
3. **Ingested files / URLs** — drop in a path or URL; Twico fetches, chunks, and embeds the content for retrieval.

Each item is shown as a collapsed preview (first 180 chars). Click any row to expand to the full content. Click the **×** to delete (confirms first).

## Folders the agent can read

Below the Knowledge list is a small "agent file access" section where you can authorise specific folders on disk. The agent uses standard `read` / `grep` tools against those paths when retrieving information — e.g. point it at your `~/notes/` directory and it can quote things back without you having to copy-paste them into Knowledge first.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://litao.gitbook.io/twico/teaching-it-about-you/brain.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.