Playmcp Kakaotalk Emoticon

🚀 Kakao Emoticon PlayMCP Server

An MCP (Model Context Protocol) server designed to automate or assist in the creation of KakaoTalk emoticons.

🚀 Quick Start

The PlayMCP server simplifies the process of creating KakaoTalk emoticons. It offers a range of features from previewing emoticon plans to generating and validating emoticons according to KakaoTalk's specifications.

✨ Features

1. before-preview (Emoticon Planning Preview)

• Used in the planning stage before emoticon creation.
• Generates a preview page in the style of a KakaoTalk chat room.
• Displays descriptive text at each emoticon position.

2. generate (Emoticon Generation)

• Generates AI emoticons using the Hugging Face API.
• Creates emoticons based on character images.
• Automatically generates emoticons if no character image is provided.
• Converts videos to WebP format for animated emoticons.

3. after-preview (Final Preview)

• Provides a preview page with actual emoticon images.
• Allows users to download a ZIP file of the emoticons.
• Enables users to zoom in on emoticons by clicking on them.

4. check (Emoticon Validation)

• Checks emoticons against KakaoTalk's submission specifications.
• Validates file format, size, capacity, and quantity.

📦 Emoticon Specifications

Static Emoticons

Dynamic Emoticons

Big Emoticons

Static Mini Emoticons

Dynamic Mini Emoticons

🔑 Hugging Face Token Authentication

To use the emoticon generation (generate tool), you need a Hugging Face API token.

Passing the Token via the Authorization Header

Pass the token as a Bearer token in the Authorization header of your HTTP request.

Authorization: Bearer hf_xxxxxxxxxxxxxxxxxxxxx

When using PlayMCP, you can securely store and automatically pass the token through PlayMCP's authentication settings.

Obtaining a Hugging Face Token

1. Log in to Hugging Face.
2. Navigate to Settings → Access Tokens.
3. Click "New token".
4. Enter a token name and set the permissions (read permission is required).
5. Securely store the generated hf_xxx... token.

📦 Installation and Execution

1. Install Dependencies

pip install -r requirements.txt

2. Set Environment Variables (Optional)

# Server settings (Optional)
export HOST="0.0.0.0"
export PORT="8000"
export BASE_URL="https://your-server-url.com"

# Redis settings (Recommended - Use memory storage if not set)
export REDIS_URL="redis://localhost:6379"
# Or if there is a password:
# export REDIS_URL="redis://:password@host:port"

⚠️ Important Note

The Hugging Face token must be passed by the user via the Authorization header or the hf_token parameter, not as an environment variable.

3. Start the Server

python server.py

Or

uvicorn server:app --host 0.0.0.0 --port 8000

🚀 Deployment on Railway

1. Create a Railway Project

1. Log in to Railway.
2. Click "New Project".
3. Select "Deploy from GitHub repo".
4. Choose this repository.

2. Set Environment Variables (Optional)

You can add the following environment variables in the Variables tab of the Railway dashboard:

⚠️ Important Note

Railway automatically sets the PORT. The Hugging Face token must be passed by the user.

3. Verify Deployment

• Railway automatically builds and deploys the application.
• Verify the deployment by accessing the provided URL: https://playmcp-kakaotalk-emoticon.bloupla.net/health

4. Register PlayMCP

After deployment, register the MCP SSE endpoint URL with PlayMCP:

https://playmcp-kakaotalk-emoticon.bloupla.net/sse

📚 PlayMCP Registration

1. Log in to PlayMCP with your Kakao account.
2. Enter the server URL in the MCP server registration menu.
3. Set up the integration with your Hugging Face account.
4. Switch to public access after testing.

💻 Usage Examples

before-preview

{
  "emoticon_type": "static",
  "title": "Cute Cat",
  "plans": [
    {"description": "Cat saying hello", "file_type": "PNG"},
    {"description": "Sleepy cat", "file_type": "PNG"}
  ]
}

generate

{
  "emoticon_type": "static",
  "character_image": "data:image/png;base64,...",
  "emoticons": [
    {"description": "Cat waving hello", "file_extension": "png"},
    {"description": "Cat sleeping with eyes closed", "file_extension": "png"}
  ]
}

after-preview

{
  "emoticon_type": "static",
  "title": "Cute Cat",
  "emoticons": [
    {"image_data": "data:image/png;base64,..."},
    {"image_data": "data:image/png;base64,..."}
  ],
  "icon": "data:image/png;base64,..."
}

check

{
  "emoticon_type": "static",
  "emoticons": [
    {"file_data": "base64...", "filename": "emoticon_01.png"}
  ],
  "icon": {"file_data": "base64...", "filename": "icon.png"}
}

🔧 Technical Details

Models Used

• Image Editing: Qwen/Qwen-Image-Edit
• Video Generation: Wan-AI/Wan2.1-I2V-14B-480P
• Character Generation: black-forest-labs/FLUX.1-schnell

📄 License

This project is licensed under the MIT License.