# Playground AI MCP for AI Agents AI Agent Connect > Playground AI. Generate, inpaint, upscale, and transform high-resolution images using natural language. This Connector connects your AI agent to Playground AI's compute clusters, allowing you to create assets, remove backgrounds, and expand image borders without leaving your primary workspace. ## Overview - **Category:** ai-frontier - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_CZ4iZgNPLcXoXLZeQcrF98W9DBO3Kd80p4W5elUg/ai-agent-connect - **Tags:** generative-art, image-generation, inpainting, outpainting, ai-models, creative-workflow ## Description Connect your AI agent directly to Playground AI's compute clusters. This means you can stop dragging files between tabs and just tell your agent what you need. Whether you want to create a new high-resolution asset from scratch or surgically fix a small area of an existing photo, this Connector handles the heavy lifting. You can generate images, remove backgrounds, and even expand your canvas using outpainting. Because Vinkius makes it easy to manage these connections, you can keep your focus on the creative side of your project. It's about moving from manual clicking to a unified workflow where your vision is just a few words away from a finished file. ## Tools ### inpaint_image Edits specific parts of an image by using a prompt and a mask image. This keeps the rest of the image identical while you fix a detail. ### list_generations Lists your most recent image generations and prompts. Use this to find previous work and IDs. ### generate_image Creates new images from text prompts using models like Playground v3 or SDXL. You can specify dimensions and style nuances. ### transform_image Modifies an existing image based on a text prompt and a strength setting. This is useful for changing the style of a base asset. ### upscale_image Increases the resolution and detail of an image by a factor of 2 or 4. Use this to turn low-res web assets into high-res prints. ### remove_background Strips the background from an image and returns a transparent PNG. It is perfect for isolating products or people for UI design. ### outpaint_image Extends an image beyond its current borders in any direction. The AI generates new content that matches the existing style. ### generate_with_controlnet Uses a reference image to guide the generation with depth, pose, or edges. This provides high-level structural control over the output. ### list_models Shows all available models and their specific capabilities. Use this to see what versions are currently active. ### get_generation Fetches the details and metadata for a specific generation ID. This helps you track the status of an image request. ## Prompt Examples **Prompt:** ``` Generate a 1024x1024 image of a cyberpunk coffee cup in neon lighting. ``` **Response:** ``` I've generated that cyberpunk coffee cup for you in 1024x1024 resolution. **Result:** [Link to Image] Would you like me to upscale it to 4x for a larger banner? ``` **Prompt:** ``` Upscale this image to 4x its size https://example.com/small_icon.png. ``` **Response:** ``` I've processed that icon through the upscaler. It's now 4x larger while keeping the details sharp. **High-Res Link:** [Link to Image] ``` **Prompt:** ``` Remove the background from the image at https://example.com/person.jpg. ``` **Response:** ``` I've stripped the background from that photo. You now have a clean, transparent PNG ready for your UI. **Asset:** [Link to Transparent PNG] ``` ## Capabilities ### Generate high-resolution images Create new images from text prompts using models like Playground v3 or SDXL. ### Remove backgrounds Strip the background from any photo to return a clean, transparent PNG. ### Upscale images Increase the resolution of blurry or small images by up to 4x. ### Expand image borders Use outpainting to intelligently grow the edges of an image in any direction. ### Edit specific image areas Use inpainting to modify a specific part of a photo while keeping the rest intact. ### Apply style guidance Use ControlNet to guide generation with depth, pose, or edge information. ## Use Cases ### Generating UI placeholders A developer needs a specific icon for a mockup. They ask the agent to generate a 1024x1024 neon coffee cup and put it straight into their project. ### Fixing low-res logos A designer has a small logo that looks blurry on a banner. They use upscale_image to make it sharp enough for the final layout. ### Isolating product photos A marketer has a product photo with a messy room background. They use remove_background to isolate the item for a clean ad. ### Iterative style sketching An artist wants to see a sketch as a polished render. They use transform_image to mutate the sketch into a final visual style. ## Benefits - Stop tab-hopping by generating assets directly within your AI client using generate_image. - Get production-ready transparent files instantly with the remove_background tool. - Fix low-quality assets by increasing resolution up to 4x with upscale_image. - Expand your canvas without manual painting by using outpaint_image to add new content. - Maintain brand consistency by applying specific styles via generate_with_controlnet. - Edit specific areas of a photo without re-generating the whole thing using inpaint_image. ## How It Works The bottom line is you get a direct line to high-end image generation without the manual friction of a web interface. 1. Add the Playground AI integration to your Connector setup. 2. Provide your Playground API key in the configuration. 3. Start requesting image generations or edits directly in your chat. ## Frequently Asked Questions **Can I use Playground AI to make my images bigger?** Yes, you can use this to upscale images up to 4x while keeping the details sharp. It is great for turning small web icons into large hero images. **Does this help with removing backgrounds from product photos?** Definitely. You can tell your agent to strip the background, and it will return a transparent PNG file ready for your website. **How do I expand the edges of an image?** You can use the outpainting feature to tell your agent which direction to grow the image, and it will fill in the new space with matching content. **Can I change parts of an image without changing the whole thing?** Yes, you can use inpainting. Just tell your agent which area to edit, and it will only modify that specific spot. **Is this good for web developers who need icons?** It is perfect for that. You can generate specific assets, like buttons or icons, directly through your chat and put them straight into your code. **Can I use different styles for my images?** Yes, the integration supports several models and styles, including ControlNet for specific depth and pose guidance. **Can my AI automatically remove a background and embed the image into my project?** Yes! The agent triggers the `remove_background` method on any public image URL. Playground internally isolates the core subject, dropping the background layers, and replies with a secure pointer URL. Your agent can instantly fetch the response string and code an `` tag placing the clean asset directly into the UI mapping you're developing. **How exactly does `outpaint_image` expand the canvas?** The tool passes the target image alongside the literal specific expansion vectors ('up', 'down', 'left', 'right') and a guiding context text. The diffusion nodes generate purely new mathematical tensor states radiating outward, predicting the lighting, shadows, and environment strictly following your prompt rules without stretching existing pixels. **Can I enforce specific structural references using ControlNet constraints?** Yes. Instead of standard generation, invoke `generate_with_controlnet`. You upload a scribble or an edges-only mapping layout as your source. By defining the explicit type ('canny', 'depth', 'pose', 'scribble'), the AI strictly anchors its conceptual render to those physical guiding constraints rather than randomizing spatial architecture.