MCP Server Stability Ai

🚀 Stability AI MCP Server for Claude Desktop

This is an MCP server for the Claude desktop application, enabling image generation and manipulation through Stability AI's API. Users can invoke models like Stable Diffusion via specific commands to achieve functions such as image generation and editing.

🚀 Quick Start

Installation

Manual Installation

1. Create a folder on your computer to store generated or modified images. Recommended paths are as follows:

   • macOS: /Users/<username>/Downloads/stability-ai-images
   • Windows: %APPDATA%\Claude\mcp-server-stability-ai\images

2. Ensure you have a Node.js environment to run the npx command. If not, download and install it from the Node.js official website.

3. Modify your claude_desktop_config.json file by adding the following content:

   {
     "mcpServers": {
       "stability-ai": {
         "command": "npx",
         "args": [
           "-y",
           "mcp-server-stability-ai"
         ],
         "env": {
           "STABILITY_AI_API_KEY": "sk-1234567890"
         }
       }
     }
   }
   

4. Restart the Claude desktop application to complete the configuration.

Installation via Smithery

Automatically install through Smithery:

npx @smithery/cli install mcp-server-stability-ai --client claude

SSE Mode

If you want to deploy the server as an HTTP service for third - party use, you can start the SSE mode:

npx mcp-server-stability-ai -y --sse

In this mode, the server needs to set the following environment variables to use Google Cloud Storage to store image files:

• GCS_PROJECT_ID
• GCS_CLIENT_EMAIL
• GCS_BUCKET_NAME
• GCS_PRIVATE_KEY

Note that the multi - tenant isolation mechanism in this mode is basic and insecure. It segments storage prefixes by the requester's IP address and makes all images publicly accessible.

✨ Features

• Text - to - Image Generation: Use the Stable Diffusion model to convert text descriptions into images.
• Image Manipulation: Includes advanced operations such as image search - and - replace (search-and-x), inpainting, etc.
• Model Support: Compatible with the latest Stable Diffusion 3.5 series models.

💻 Usage Examples

Basic Usage

Text - to - Image Generation

Use the Stable Diffusion model to convert text descriptions into high - quality images. For example:

/stability-ai generate --prompt "A flying owl"

Image Manipulation

Search - and - Replace (search-and-x)

Modify image content by specifying search and replacement terms:

/stability-ai search-and-replace --prompt "A flying owl" --negative_prompt "Sky"

Inpaint

Repair or modify an image, requiring a mask for the repair area:

/stability-ai inpaint --image_path "path/to/image.jpg" --mask_path "path/to/mask.png"

Advanced Usage

The server is compatible with the latest Stable Diffusion 3.5 series models, including but not limited to:

• SD 3.5 Base
• SD 3.5 Fine - tuned for images
• SDXL

📚 Documentation

Environment Variable Configuration

When the server is running, you can customize its behavior by setting the following environment variables:

Logging and Monitoring

The server outputs detailed logs during operation, facilitating debugging and monitoring. Log contents include:

• Request processing time
• Operation type
• Error information (if any)

Tutorials and Videos

If you prefer video tutorials, you can refer to the following link:

How to Configure and Use the Stability AI MCP Server

Troubleshooting

Common Issues

1. Unable to connect to the Stability AI API

   • Ensure you have correctly set the STABILITY_AI_API_KEY.
   • Check your network connection to ensure you can access external services.

2. Image generation fails

   • Check the log file for specific error information.
   • Ensure that the input prompt description is clear and meets the model requirements.

3. Performance issues

   • Ensure your hardware configuration is sufficient (GPU acceleration is recommended).
   • Optimize generation parameters, such as reducing the resolution or the number of iterations.

📄 Changelog

Version 1.0.0

• First release, supporting basic text - to - image generation and search - and - replace functions.
• Compatible with the Stable Diffusion 3.5 series models.

Version 1.1.0

• Added the inpaint function.
• Supported Google Cloud Storage integration.
• Optimized the logging system to improve debugging efficiency.

Version 1.2.0

• Enhanced the security of the SSE mode.
• Improved the multi - tenant isolation mechanism.
• Improved overall performance and stability.

🤝 Contribution and Support

Contribute Code

The community is welcome to contribute code! Please visit the GitHub repository for detailed contribution guidelines.

Get Support

If you need help or encounter problems, please contact:

• Email: support@stability.ai
• Forum: Stability AI Community Forum

📄 License

This project is licensed under the MIT License. Please refer to the accompanying LICENSE file for specific terms.