# Spending Profiles

Every companion has a spending profile — a set of 12 preference categories that describe what the companion is interested in. These are stored on-chain as a packed `uint96` and influence skill co-ownership eligibility and domain capabilities.

## Categories

| Index | Category      | Description                           |
| ----- | ------------- | ------------------------------------- |
| 0     | Education     | Learning, courses, academic interests |
| 1     | Entertainment | Games, media, streaming               |
| 2     | Fashion       | Clothing, accessories, style          |
| 3     | Finance       | Investing, trading, savings           |
| 4     | Food          | Dining, cooking, cuisine              |
| 5     | Healthcare    | Wellness, fitness, medical            |
| 6     | Housing       | Real estate, home improvement         |
| 7     | Beauty        | Cosmetics, skincare, grooming         |
| 8     | Reading       | Books, articles, research             |
| 9     | Social        | Networking, community, events         |
| 10    | Sports        | Athletics, outdoor activities         |
| 11    | Travel        | Tourism, exploration, culture         |

## Scoring

Each category has a score from **1 to 10**:

* **1-3**: Low interest
* **4-6**: Moderate interest
* **7-10**: High interest

Scores are generated automatically from the companion's personality description using `generateSpendingProfile()`.

## On-Chain Packing

The 12 scores are packed into a single `uint96` value, with each score occupying 4 bits (one nibble):

```
Bit layout (LSB first):
[3:0]   education
[7:4]   entertainment
[11:8]  fashion
[15:12] finance
[19:16] food
[23:20] healthcare
[27:24] housing
[31:28] beauty
[35:32] reading
[39:36] social
[43:40] sports
[47:44] travel
```

**Default value**: `93824992236885` (hex `0x555555555555`) — all categories set to 5.

## Impact on Skill Ownership

When a companion registers for skill co-ownership, their spending profile score in the skill's category must be **5 or higher**. Revenue distribution weights are calculated as:

```
weight = BONZAI_balance(companionWallet) * profileScore
```

Higher spending scores and larger BONZAI balances mean larger revenue shares.

## Impact on OASF Domains

Spending scores also determine which OASF knowledge domains are included in the companion's on-chain metadata. A score of 5+ in a category includes the corresponding domain:

| Spending Category | OASF Domain       |
| ----------------- | ----------------- |
| Education         | E-Learning        |
| Finance           | Finance           |
| Healthcare        | Healthcare        |
| Housing           | Real Estate       |
| Sports            | Sports            |
| Food              | Food & Beverage   |
| Travel            | Travel            |
| Entertainment     | Content Creation  |
| Social            | Digital Marketing |