home / mcp / ghibli mcp server
Provides an MCP server to generate Ghibli-style images via an OpenAI-based API with SDL-based communication.
Configuration
View docs{
"mcpServers": {
"bach-ai-tools-bachai-ghibli-image-generator-api-open-ai-4o-image-generation-free": {
"url": "https://mcp.example.com/mcp",
"headers": {
"API_KEY": "your_api_key_here"
}
}
}
}You can access the Ghibli Image Generator API through an MCP server that runs locally or via a managed runtime. This setup lets you generate Ghibli-style images by calling a dedicated tool from your application, with simple environment-based authentication and a lightweight transmission protocol.
You will use an MCP client to connect to a stdio-based server that runs the Ghibli image generator. First obtain your API key and place it in your environment so the client can authorize requests. Then start the MCP server via the provided runtime (uvx) and point your client to the local server endpoint exposed by the MCP runtime. You can call the image generation tool through the MCP interface and receive generated images as responses. The server exposes a dedicated tool named ghibli_generate_image for image creation and a companion endpoint for response handling.
Your client should supply the image generation request payload to the MCP server using the supported tool name and endpoint. Ensure the API key is set in the environment where the MCP server runs. Monitor usage in real time through your MCP management interface if available.
Prerequisites you need before installation:
- Python with pip
- An MCP-compatible environment or manager (such as UVX) to run stdio MCP servers
- An API key for the Ghibli image generation serviceStep-by-step commands you should run in order.
pip install bach-ghibli_image_generator_api_open_ai_4o_image_generation_free
# Or install in editable mode from source (if you have the source checked out)
pip install -e .
# Then run the server locally via UVX (recommended) or a direct Python command
uvx --from bach-ghibli_image_generator_api_open_ai_4o_image_generation_free bach_ghibli_image_generator_api_open_ai_4o_image_generation_free
# Alternatively, run directly in development mode
python server.py
# If you want to run as a command after install
pip install bach-ghibli_image_generator_api_open_ai_4o_image_generation_free
bach_ghibli_image_generator_api_open_ai_4o_image_generation_free
# Ensure the API key is set in the environment before starting
export API_KEY="your_api_key_here"API authentication is required. Set your API key in the environment where the MCP server runs.
export API_KEY="your_api_key_here"{
"mcpServers": {
"bach-ghibli_image_generator_api_open_ai_4o_image_generation_free": {
"command": "uvx",
"args": ["--from", "bach-ghibli_image_generator_api_open_ai_4o_image_generation_free", "bach_ghibli_image_generator_api_open_ai_4o_image_generation_free"],
"env": {
"API_KEY": "your_api_key_here"
}
}
}
}The MCP server provides a tool named ghibli_generate_image to request image generation from the Ghibli API. The server exposes a dedicated endpoint for the image generation action so you can generate images by calling the tool through your MCP client.
Keep your API key secure. Do not commit keys to code repositories. Use environment variables and secret management where possible. Rotate keys periodically and monitor usage for any unusual activity.
If you cannot connect to the MCP server, verify that the UVX runtime is installed and that the environment variable API_KEY is set in the session running the server. Check that the command and arguments mirror the examples shown in the configuration blocks.
Ghibli generate tool to request image generation from the API via the MCP endpoint.
Example response handler tool for processing ghibli image responses.