Scenario AI

🚀 MCP Server Project Based on Scenario.com

This project is an MCP server based on Scenario.com, offering image generation and background removal capabilities. Users can run this server in a development environment by configuring the API key and model ID.

🚀 Quick Start

Repository Structure

The project's file structure is as follows:

• src/requirements.txt: A list of project dependencies.
• src/scenario_mcp_server.py: The main program of the MCP server.
• examples/client_example.py: Example client code for using the MCP server.
• scripts/start_server.bat/sh: Scripts to start the MCP server (compatible with Windows/Linux/MacOS).
• scripts/run_client_example.bat/sh: Scripts to run the example client.
• scripts/run_tests.bat/sh: Scripts to execute the test suite.

Setup Steps

1. Install Python Environment: Ensure that Python 3.6 or a higher version is installed on your system. It is recommended to use a virtual environment to manage dependencies.
2. Create and Activate a Virtual Environment:

   python -m venv .venv
   source .venv/bin/activate  # MacOS/Linux
   .\\.venv\\Scripts\\activate  # Windows
   

3. Install Dependencies: Install the required Python packages in the virtual environment.

   pip install -r src/requirements.txt
   

4. Configure Environment Variables: Create and edit the config.env file, adding the following content:

   SCENARIO_API_KEY=your_api_key
   SCENARIO_API_SECRET=your_api_secret
   SCENARIO_MODEL_ID=model_id
   

5. Install MCP SDK and python-dotenv:

   pip install mcp python-dotenv
   

Running the Server

Start the MCP Server

Use the following commands to start the server:

# MacOS/Linux
mcp-server src/scenario_mcp_server.py --config ./.env

# Windows
set SCENARIO_API_KEY=your_api_key & set SCENARIO_API_SECRET=your_api_secret & set SCENARIO_MODEL_ID=model_id & mcp-server src\scenario_mcp_server.py --config .\.env

✨ Features

Available Tools

Tool List

1. Generate an Image:

   mcp generate-image --prompt your_prompt_here [--negative-prompt your_negative_prompt] [--num-samples 5]
   

2. Remove Background:

   mcp remove-background --asset-id your_asset_id
   

Resources

1. Get Server Status Information: status://info
2. Query Job Status: job://{job_id}
3. View Asset Information: asset://{asset_id}

💻 Usage Examples

Basic Usage

In examples/client_example.py, users can find example code for calling the MCP server. Run the script using the following commands:

# MacOS/Linux
python examples/client_example.py

# Windows
python examples\client_example.py

📚 Documentation

Script Usage

• start_server.bat/sh: Start the MCP server.
• run_client_example.bat/sh: Run the example client.
• run_tests.bat/sh: Execute the test suite.

Testing and Viewing Documentation

Run the following command to execute tests and view documentation:

python -m pytest tests/test_server.py -v

The test file tests/test_server.py details how to use the MCP server on Scenario.com, including:

1. Server configuration instructions.
2. Available tools and their parameters.
3. Resource access methods.
4. A complete workflow example using the Roo - Cline interface.

🔧 Troubleshooting

Issues Starting the Server

1. Ensure that the MCP SDK and python-dotenv are installed:

   pip install mcp python-dotenv
   

2. Check if all dependencies are included in the virtual environment:

   pip install -r src/requirements.txt
   

3. Verify that the config.env file is correctly configured.

API Errors

1. Confirm that the API key and secret are correct.
2. Verify the validity of the model ID.
3. Refer to the official Scenario.com documentation for the latest information.

📄 License

This project is licensed under the MIT License. Please refer to the LICENSE file for detailed information.

By following the above steps, users can successfully configure and run the MCP server based on Scenario.com to achieve image generation and background removal functions.