# Stability AI MCP > Stability AI Alternative MCP connects your agent to professional-grade image generation and editing capabilities using Stable Diffusion models. Generate stunning visuals from simple text prompts, modify existing photos, or upscale low-res images—all directly through conversation with any compatible client. ## Overview - **Category:** ai-frontier - **Price:** Free - **Tags:** stable-diffusion, image-generation, ai-art, upscaling, inpainting ## Description Need high-quality visuals without leaving your chat window? This MCP gives your AI agent direct access to the power of Stability AI. You can generate entirely new art just by describing it. Need to change an existing photo, like replacing a jacket or fixing a background? The system handles that. Plus, if you have great images but they're too small, you can increase their resolution and detail. It’s all about getting visual assets quickly, whether for a blog post or a design concept. Because this MCP is housed in the Vinkius catalog, you connect once from your favorite AI client (like Claude or Cursor) and get access to image creation alongside everything else you do. ## Tools ### get_user_account Retrieves your current account details and usage information from Stability AI. ### list_engines Lists all the specific Stable Diffusion models and upscaling engines available for use in image generation. ### masking Replaces parts of an image you provide by defining a precise mask over the area to be changed. ### get_user_balance Checks your remaining credit balance, so you know exactly how many generations you can afford. ### image_to_image Modifies an existing picture based on a new text prompt, guiding the AI to change its content while keeping the structure intact. ### text_to_image Generates a completely new image from scratch using only descriptive text prompts. ### upscale_image Increases the resolution of an existing picture, making it clearer and more detailed for printing or large displays. ## Prompt Examples **Prompt:** ``` Generate an image of a cyberpunk city at night with neon lights using stable-diffusion-xl-1024-v1-0. ``` **Response:** ``` I've initiated the generation using the Stable Diffusion XL engine. Here is your generated image of a neon-lit cyberpunk city. ``` **Prompt:** ``` Check my current Stability AI credit balance. ``` **Response:** ``` Your current account balance is 145.5 credits. ``` **Prompt:** ``` List all available engines for image generation. ``` **Response:** ``` I've retrieved the list of available engines. You can use models like `stable-diffusion-v1-6`, `stable-diffusion-xl-1024-v1-0`, and upscalers like `esrgan-v1-x2plus`. ``` ## Capabilities ### Generate images from text prompts Creates brand-new pictures based on a descriptive text prompt, returning the image data directly to your agent. ### Modify existing photos Alters an image you provide using a new text prompt, letting you change elements like style or subject matter. ### Mask and replace specific areas of images Allows precise editing by defining a mask over an image and replacing only the masked area with something else. ### Increase image resolution Upscales low-resolution pictures, adding detail and boosting overall clarity without losing quality. ### Check account status and engines Provides real-time data on your available credits and lists all the specific Stable Diffusion models you can use for generation. ## Use Cases ### Revamping a blog header image A marketing manager needs a new hero image for an article about space travel. They prompt their agent: 'Generate an astronaut looking at Saturn.' The agent uses `text_to_image` to create the perfect visual, saving hours of searching stock photo sites. ### Correcting a client's logo draft A designer gets a rough concept drawing. They upload it and tell their agent: 'Keep this layout but make the background look like distressed wood.' The agent uses `image_to_image` to execute the precise style modification. ### Scaling artwork for print An artist finishes a beautiful 1080p piece. Before sending it to a printer, they use the MCP's `upscale_image` tool. The agent increases the resolution dramatically so the final product looks professional and crisp. ### Updating character details in concept art A comic book writer needs to change a character’s hairstyle mid-project. They use `masking`, marking over the hair area, and prompt: 'Make it shoulder-length blonde.' The MCP isolates the change perfectly. ## Benefits - Stop leaving your chat client to create art. With this MCP, you can generate images (using `text_to_image`) or edit photos using only conversational prompts, keeping your workflow in one place. - You don’t have to guess about model compatibility. Use `list_engines` to see every available generation model and upscaler before you commit a credit. - When an image isn't quite right, use `image_to_image`. This tool lets you guide the AI to modify existing concepts—like changing a character’s clothing or background—without starting over. - Need surgical precision? The `masking` tool lets you replace only specific sections of a photo. You mark the area, and your agent handles the complex replacement process. - Don't worry about low-res assets. Run `upscale_image` on any generated piece to boost its resolution, making it suitable for high-quality print or large screens. ## How It Works The bottom line is: it turns complex, multi-step image processing into a simple conversation with your agent. 1. First, connect this MCP to your preferred AI client and enter your Stability AI API key. 2. Next, tell your agent exactly what kind of image you want or how you want to modify a photo. The agent selects the right tool based on your request. 3. Finally, the MCP runs the generation process, returning the resulting image data (base64 encoded) for you to review and use. ## Frequently Asked Questions **How do I check my Stability AI credit balance using the MCP?** You use the `get_user_balance` tool. Simply ask your agent to 'Check my account balance.' It will pull up the real-time number, letting you know exactly how many generations are left. **What is the difference between `text_to_image` and `image_to_image`?** `text_to_image` creates something entirely new from a text description. `image_to_image` starts with an existing photo and modifies it based on your prompt, keeping some of the original composition. **Can I use this MCP to change one small part of a picture?** Yes, you'll want to use the `masking` tool. You first define the area you want to change with a mask, and then prompt your agent for the replacement content. **Does this MCP support upscaling images for print?** Yes, you can use `upscale_image`. This function increases the resolution of an existing picture so it looks sharp even when printed large or viewed on a massive screen.