> ## Documentation Index
> Fetch the complete documentation index at: https://docs.alana.shopping/llms.txt
> Use this file to discover all available pages before exploring further.

# Brands

> Manage multiple brands within a workspace, each with its own AI personality.

## Overview

A **brand** represents a distinct product line or company within your workspace. Each brand carries its own identity, logo, and AI specifications that shape how content is generated.

## Brand properties

| Property      | Description                             |
| ------------- | --------------------------------------- |
| `name`        | Brand display name                      |
| `slug`        | URL-safe identifier                     |
| `logo_url`    | Brand logo (stored in Supabase Storage) |
| `description` | Brief brand description                 |

## AI Specifications

Every brand can define an **AI spec** — a set of guidelines that the AI engine follows when generating or enhancing content for that brand.

| AI Spec Field          | Purpose                                                     |
| ---------------------- | ----------------------------------------------------------- |
| `tone_of_voice`        | How the brand sounds (e.g. "professional yet approachable") |
| `brand_objective`      | What the brand aims to achieve with its content             |
| `target_audience`      | Who the content is for                                      |
| `content_policies`     | Rules the AI must follow (e.g. "never mention competitors") |
| `language_preferences` | Preferred language and regional variations                  |
| `keywords_always`      | Words that should appear in generated content               |
| `keywords_never`       | Words the AI must avoid                                     |
| `custom_instructions`  | Free-form additional guidance                               |

<Tip>
  The more detailed your AI specs, the better the generated content matches your brand voice. Invest time in `tone_of_voice` and `content_policies` for the best results.
</Tip>

## Brand-aware AI

When you call any AI endpoint (`/ai/chat`, `/ai/generate`, `/ai/enhance`) with a `brandId`, the engine automatically:

1. Loads the brand's AI specs
2. Injects them into the system prompt
3. Retrieves relevant product context via RAG
4. Generates content that respects tone, keywords, and policies

This means two brands in the same workspace can produce completely different content styles from the same prompt.