# autoRetouch MCP for AI Agents AI Agent Connect > autoRetouch lets you automate high-fidelity image retouching for e-commerce. It handles background removal, color correction, and beauty editing at scale. You can manage entire batches of product photos, track processing statuses, and monitor your wallet balance through your preferred AI client. ## Overview - **Category:** ecommerce - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_5AWw7Y7S7Xbt5L7ntz9BdW1fFNyLMk1GJi8fnC6W/ai-agent-connect - **Tags:** photo-editing, background-removal, image-processing, automation, e-commerce-tools, retouching ## Description autoRetouch lets you manage a catalog with thousands of product photos without the manual headache. Right now, you're likely moving files into folders, opening web tools, and manually clicking buttons to remove backgrounds or fix colors. It's a slog. With this Connector, your AI agent becomes a digital imaging engineer. You can tell your agent to start a batch of images for a new collection, and it handles the heavy lifting of sending them to the platform for processing. You don't have to watch the progress bars; the agent can check the status and let you know when the high-fidelity results are ready for download. This setup turns a manual production line into a conversation. Because Vinkius hosts this Connector, you can plug it into your existing workflow and keep your creative focus on the big picture rather than the individual pixels. You get a coordinated media pipeline where your agent handles the uploads, the batch tracking, and the final delivery of URLs. It's about moving from 'doing the work' to 'directing the work.' ## Tools ### create_execution Start an image processing task to apply effects like background removal. You get a unique ID to track the progress of the edit. ### get_execution Pull the current status and metadata for a single image processing task. Use this to see exactly when a specific photo is ready. ### get_image Retrieve specific details for an individual image in your library. This is useful for checking the properties of a single asset. ### get_organization Get your organization's profile details and account info. Use this to verify your account settings and organizational data. ### get_wallet_balance Check how much credit you have left in your autoRetouch wallet. This keeps you informed of your remaining budget for image processing. ### get_workflow See the specific parameters of a saved image processing workflow. This allows you to review the exact steps used for a particular edit. ### list_batches See a full list of all batches you've created. This makes it easy to manage multiple sets of product photos at once. ### list_executions View a history of your recent image processing tasks. This helps you keep track of every edit your agent has performed. ### list_images See all the images you've uploaded to your account. This provides a quick way to browse your entire media library. ### list_workflows Browse all the image processing workflows available in your account. This lets you quickly find and select the right editing pipeline. ### get_batch Fetch the specific details and status of a previously created batch. This helps you manage large groups of images as a single project. ## Prompt Examples **Prompt:** ``` List all active image processing workflows in my organization. ``` **Response:** ``` I've retrieved your workflows. You currently have 3 active AI pipelines: **Ghost Mannequin** (ID: wf_123), **Shadow Gen**, and **Bulk Background Remove**. Which one should we use? ``` **Prompt:** ``` Run the 'Ghost Mannequin' workflow (ID: 'wf_123') on image ID 'img_456'. ``` **Response:** ``` Execution triggered! I've started processing image img_456 using the specified workflow. The execution ID is **exec_789**. I'll monitor the status and notify you when the high-fidelity result is ready. ``` **Prompt:** ``` Check status and get the result URL for execution 'exec_789'. ``` **Response:** ``` Fetching status... Execution exec_789 is currently **Success**. You can access your high-fidelity processed image here: [result_url]. Need any help with more batch tasks? ``` ## Capabilities ### Start high-fidelity image processing Trigger automated edits like background removal and shadow generation for your assets. ### Check task status Get real-time updates on whether a specific image is still processing or finished. ### Retrieve processed URLs Get the final web links for your finished images as soon as they are ready. ### Group images into batches Organize multiple image tasks into a single tracked group for easier management. ### List active workflows See all the custom image processing pipelines available in your account. ### Monitor wallet balance Check your remaining account credits directly through your AI client. ## Use Cases ### Catalog Refresh An e-commerce owner tells their agent to 'Remove the background from all 50 new sneaker photos' and the agent uses create_execution to start the batch. ### Agency Scaling A marketing lead asks the agent to 'Apply the Ghost Mannequin workflow to the latest clothing set,' and the agent handles the batch processing. ### Photographer Workflow A photographer wants to check progress on a shoot and asks, 'Are the lifestyle shots finished yet?' The agent checks the status and provides the URLs. ### Budget Monitoring An ops manager asks, 'How much credit do we have left for the holiday campaign?' and the agent pulls the balance using get_wallet_balance. ## Benefits - Manage high-volume catalogs faster by using create_execution to start retouching tasks for hundreds of photos at once. - Keep track of your production pipeline with list_batches and list_executions to see exactly what's finished and what's still in the queue. - Save time on repetitive edits by using list_workflows to select and run specific high-fidelity pipelines like shadow generation. - Stay on top of your costs by using get_wallet_balance to check your remaining credits during high-volume marketing campaigns. - Get your assets into your site faster by using get_execution to grab the final result URLs as soon as the AI finishes the work. ## How It Works The bottom line is you turn your AI client into a production-ready image editing suite that handles the tedious stuff for you. 1. Get your API Token and Organization ID from the autoRetouch dashboard. 2. Connect the Connector to your AI client using those credentials. 3. Ask your agent to retouch images, create batches, or check statuses. ## Frequently Asked Questions **Can the autoRetouch MCP handle thousands of images at once?** Yes, it's designed for high-volume processing. You can group images into batches and have your agent manage the entire queue, making it ideal for large e-commerce catalogs. **What kind of edits does the autoRetouch MCP perform?** It handles high-fidelity tasks like background removal, shadow generation, color correction, and beauty editing. You can choose specific workflows to match your brand's style. **How do I see how much credit I have left with autoRetouch?** You can just ask your AI client to check your wallet balance. It will pull the current data from your account so you can monitor your spending in real-time. **Can I use the autoRetouch MCP for professional photography?** Absolutely. It's a great way for photographers to orchestrate complex retouching sequences for lifestyle or product shoots without doing the manual work. **Does the autoRetouch MCP work with my current AI client?** Yes, it works with any MCP-compatible client like Claude or Cursor. Once connected, your agent can trigger any of the available image processing tools. **How does the autoRetouch MCP handle batch tracking?** It lets you group multiple executions into tracked batches. This means you can oversee your organization's entire image editing volume in one place. **How do I find my autoRetouch API credentials?** Log in to your account, navigate to the API section to generate an Access Token (Bearer), and find your Organization ID in the organization settings. **Can I process multiple images at once?** Yes! You can trigger individual executions programmatically and group them into a single batch for high-fidelity tracking. **What happens if a process fails?** The `get_execution_status` tool will return high-fidelity error metadata to help you diagnose and re-run the task programmatically.