> For the complete documentation index, see [llms.txt](https://docs.pletor.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.pletor.ai/build-your-system/brain.md).

# Brain

One brand, one Brain. Give your system a memory.

A Brain is your brand's living memory on Pletor. It holds everything your projects need to produce relevant on-brand work: identity, guidelines, products, knowledge and mode, structured so that AI can actually use it.

You set it up once. Every system that uses it then produces work that looks and sounds like your brand.

Your Brains live at [app.pletor.ai/brains](https://app.pletor.ai/brains). 

<figure><img src="https://2795436344-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fnyc5C4KVCklpMy0pLa15%2Fuploads%2FQaw1Dfzwjob9VyRHfNrC%2F2026-09-24%2015.29.38.gif?alt=media&amp;token=1cb36170-017f-4954-a831-2c7020310b91" alt=""><figcaption></figcaption></figure>

***

## What a Brain holds

Everything inside a Brain is a **Memory**: one concept, written as short prose, optionally illustrated with images, and nested under other Memories. Every Brain starts with the same three groups:

<table><thead><tr><th width="160">Group</th><th>What goes there</th></tr></thead><tbody><tr><td><strong>Identity</strong></td><td>Who the brand is: positioning, audience, values, brand kit (logos, color palettes, typefaces)</td></tr><tr><td><strong>Guidelines</strong></td><td>How the brand shows up: voice and tone, visual do's and don'ts, house rules, hard constraints</td></tr><tr><td><strong>Knowledge</strong></td><td>What the brand knows: products, campaigns, markets, claims, winning creatives</td></tr><tr><td><strong>Products</strong></td><td>What the brand sells: SKUs, packshots, specs, variants, pricing</td></tr><tr><td><strong>Characters</strong></td><td>Who appears in the brand's world: recurring models, mascots, spokespeople, reference shots</td></tr></tbody></table>

***

## Set up your Brain

#### Step by step

{% @arcade/embed flowId="XvfLVs9jxS4ev3zpnwnj" url="<https://app.arcade.software/share/XvfLVs9jxS4ev3zpnwnj>" %}

1. **Create a Brain.** Go to [Brains](https://app.pletor.ai/brains) and click **New Brain**.
2. **Answer a few questions.** What kind of brand this is, what you sell, where you show up. This shapes the Brain around your brand before anything is written.
3. **Feed it.** Paste your website URL, drop your files (brand book, guidelines, past campaigns, product sheets, logos), add any notes that aren't in a file, then click **Build the brain**.
4. **Watch it encode.** The Brain reads your sources (seconds for a document, a few minutes for a website) and writes Memories into the Map. Nothing is written from memory: every Memory is grounded in material you gave it.
5. **Review and refine.** Approve what the Brain proposes, give feedback on anything that's off, or edit a Memory directly. Open **Map** for the big picture, **Memories** for the file tree.

You can keep feeding your Brain any time from its chat: drop a new campaign, a competitor screenshot or a revised guideline and it gets sorted into the right place. Anything it can't read, it tells you and carries on with the rest.

### What you can upload

Public websites (crawled page by page), PDF and Word documents, images, video and audio up to 10 minutes, plain-text files (Markdown, TXT, CSV, HTML, JSON), zip archives of any of those, and plain chat.

Not read yet: spreadsheets (.xlsx) and slide decks (.pptx), so export them to CSV or PDF first. Links behind a sign-in (Figma, Canva, Google Docs, Notion, Drive) can't be crawled either: export the content as images or a PDF and drop it in. Fonts are added directly in the Brand kit's typography section.

***

## Use your Brain

Once a Brain exists, it shows up wherever you create on Pletor:

* **In Studio.** Open the **Brain** panel in the left sidebar. Browse its **Memories** and **Assets**, then drag a Memory onto the canvas: it lands as a node synced live with the Brain. Edit the Memory in the Brain and every flow using it picks up the change. Import assets the same way.
* **In chat.** Pick your active Brain and describe what you want. The chat creates from your Brain's identity, guidelines and references.
* **In Apps.** A flow built on Brain nodes carries its brand context into the app. Whoever runs the app gets on-brand results without knowing where the context comes from. See [Apps](/automate/apps.md).
* **In Batch.** Brand context applies to every row, so a whole catalog stays on-brand without per-row setup. See [Batch](/automate/batch.md).
* **From Claude and other MCP clients.** The Pletor MCP can list your Brains, summarise them, read and write Memories. Ask your assistant to "set up my brand's Brain from our website" and it does the encoding for you. See [Pletor MCP](/automate/pletor-mcp.md).

### Brain or Brand nodes?

[Brand nodes](/build-your-system/nodes/brand-nodes.md) are static: the context you type or upload into them lives in that one flow. A Brain is shared and synced: set it up once, use it in every flow and chat, update it in one place. Use Brand nodes for a one-off tweak; use a Brain for anything your whole team should stay consistent on.

***

## Demo Brains

Not sure what a well-built Brain looks like? The **Demo Brains** tab in [Brains](https://app.pletor.ai/brains) shows example Brains built by the Pletor team. Open one to see how identity, guidelines and knowledge are laid out before building your own.

***

## Plans

Brain comes with every paid plan. The number of Brains you can hold at once depends on your plan:

<table><thead><tr><th width="200">Plan</th><th>Brains</th></tr></thead><tbody><tr><td>Starter</td><td>1</td></tr><tr><td>Builder</td><td>Up to 3</td></tr><tr><td>Studio</td><td>Up to 10</td></tr><tr><td>Enterprise</td><td>Custom</td></tr></tbody></table>

***

## Tips

* **Show, don't tell.** Real examples (your actual logo files, past ads, sample copy) produce better Memories than descriptions of them.
* **Start with what you have.** Your website plus one brand document is enough for a first pass. Fill the gaps the Brain asks about afterwards.
* **Keep hard rules exact.** A legal claim, a banned word or a mandatory disclaimer should be written into the Brain verbatim, not paraphrased.
* **One Brain per brand.** Mixing two brands in one Brain confuses every flow that reads it. Agencies: one Brain per client.