Metadata-Version: 2.4
Name: bria-mcp
Version: 1.0.0
Summary: Bria.ai MCP
Project-URL: Homepage, https://bria.ai
Project-URL: Documentation, https://docs.bria.ai/mcp-authentication
Author: Bria.ai
Maintainer-email: omri <omriassa@bria.ai>
License: MIT
License-File: LICENSE
Keywords: bria,mcp
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Multimedia :: Graphics
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Requires-Python: >=3.10
Requires-Dist: bria-client>=0.1.1
Requires-Dist: mcp>=1.26.0
Description-Content-Type: text/markdown

# Bria MCP Server

An [MCP](https://modelcontextprotocol.io/) server that gives AI assistants access to [Bria's](https://bria.ai) image and video generation and editing APIs.

Built on Bria's commercially-licensed generative AI, trained exclusively on licensed data.

## Quick Start

### 1. Get an API Token

Sign up at [bria.ai](https://bria.ai) and grab your API token. See the [authentication docs](https://docs.bria.ai/mcp-authentication) for details.

### 2. Set Your API Token

```bash
export BRIA_API_TOKEN=<your-token>
```

### 3. Run

The fastest way to run is with [`uvx`](https://docs.astral.sh/uv/) (no install needed):

```bash
uvx bria-mcp
```

Or install with pip / uv and run directly:

```bash
# with pip
pip install bria-mcp

# or with uv
uv pip install bria-mcp

# then run
bria-mcp
```

### 4. Configure Your MCP Client

<details>
<summary><strong>Claude Desktop / Claude Code</strong></summary>

Add to your Claude config (`claude_desktop_config.json`):

```json
{
  "mcpServers": {
    "bria": {
      "command": "uvx",
      "args": ["bria-mcp"],
      "env": {
        "BRIA_API_TOKEN": "<your-token>"
      }
    }
  }
}
```

</details>

<details>
<summary><strong>Cursor</strong></summary>

Add to your Cursor MCP config (`.cursor/mcp.json`):

```json
{
  "mcpServers": {
    "bria": {
      "command": "uvx",
      "args": ["bria-mcp"],
      "env": {
        "BRIA_API_TOKEN": "<your-token>"
      }
    }
  }
}
```

</details>

## Available Tools

### Image Editing

| Tool | Description |
|---|---|
| `remove_background` | Remove the background from an image |
| `blur_background` | Apply blur to an image background (intensity 1-5) |
| `crop_out_foreground` | Remove background and crop tightly around the foreground |
| `increase_resolution` | Upscale an image by 2x or 4x |
| `expand_image` | Expand an image or change its aspect ratio using generative AI |
| `generate_background` | Generate a new background from a text prompt or reference images |
| `erase_foreground` | Erase the foreground object and generate the area behind it |
| `enhance_image` | Enhance detail, sharpness, and clarity of an image |

### Image Generation

| Tool | Description |
|---|---|
| `generate_image` | Generate images using Bria's FIBO model — text-to-image or image-to-image. Supports `FIBO` and `FIBO_LITE` model variants. |

### Video Editing

| Tool | Description |
|---|---|
| `remove_video_background` | Remove background from a video |
| `increase_video_resolution` | Upscale video resolution by 2x or 4x |
| `video_mask_by_prompt` | Create a video mask guided by a text prompt |
| `erase_from_video` | Erase an object from a video using a mask or text prompt |

All image tools accept a URL (recommended), file path, or base64-encoded data as input.

## Examples

> Remove the background from this product photo: https://example.com/photo.jpg

> Generate an image of a minimalist workspace with natural lighting

> Take this image and change its aspect ratio to 16:9

> Upscale this video to 4x resolution

## Requirements

- Python 3.10+
- A [Bria API token](https://docs.bria.ai/mcp-authentication)

## License

MIT — see [LICENSE](LICENSE) for details.