Skip to content

Cutout Labs changelog

com.cutoutlabs/mcp

2 changes recorded. MCP Harbor re-reads this listing from the official registry every six hours and asks the live server what it exposes, and keeps what changed between one look and the next.

  • llms.txt +17 −8

    https://www.cutoutlabs.com/llms.txt

    − Cutout Labs is a free, standalone web background remover and `/v1.0/removebg`-compatible HTTP API. It removes image backgrounds up to 16 megapixels in the browser with no account, no watermark, 10-image ZIP batch processing, interactive Erase/Restore brushes, and solid color, transparent PNG, JPG, or WebP exports.− - [remove.bg moves to Canva: choosing your web workflow](https://www.cutoutlabs.com/remove-bg-canva): remove.bg's background removal moves into Canva on 1 Dec 2026. When Canva fits, when a standalone remover fits, and how to test both with your own images.− - **Migration from remove.bg**: Drop-in compatible request format ahead of the December 1, 2026 standalone remove.bg shutdown.− - [Cleaner backgrounds for product photos](https://www.cutoutlabs.com/product-photo-backgrounds): Prepare product photos for background removal, inspect fine edges, and export consistent listing images without changing your product's appearance.− - [Change Image Background Color Online (White, Custom Hex & PNG)](https://www.cutoutlabs.com/change-background-color): Change any image background color online for free. Replace photo backgrounds with pure white (#FFFFFF), custom hex colors, or gradients and download full-res PNGs.− - [Remove Background Around Hair & Fine Fur Edges Online](https://www.cutoutlabs.com/remove-background-from-hair): Remove portrait backgrounds around flyaway hair, curls, and fur while keeping soft alpha transparency, natural wisps, and clean edges without halos.− - [Client-side mask upscaling for 16 MP cutouts](https://www.cutoutlabs.com/high-resolution-background-removal): See why uploading a 25 MB photo is unnecessary for AI segmentation, and how applying a 2.5 MP alpha mask to your local canvas preserves 16 MP pixel detail.− - **Which formats can I download?** PNG keeps transparency. JPG and WebP are smaller; JPG always uses a solid background. You can also copy the cutout straight to your clipboard.+ Cutout Labs offers a free web background remover, an HTTP API, and a Model Context Protocol (MCP) server. The web editor sends an image or a resized proxy to the server for background removal; brush edits, background compositing, and exports happen in your browser. The web editor supports working images up to 16 megapixels, no account, no watermark, and batches of up to 10 images. The API requires a key and has separate input and quota limits.+ ## Docs+ - [Full documentation](https://www.cutoutlabs.com/llms-full.txt): API reference, MCP setup, and every guide in one plain-text file+ - **Migration from remove.bg**: Compatible request fields and endpoint path, with differences including a 12 MB / 12 MP input limit, no automatic foreground-type detection, and no car or 3D shadows. Review the [API reference](https://www.cutoutlabs.com/background-removal-api) and test your existing parameters before switching.+ - [Change image background color online (white, custom hex & PNG)](https://www.cutoutlabs.com/change-background-color): Change any image background color online for free. Replace photo backgrounds with pure white (#FFFFFF), custom hex colors, or gradients and download full-res PNGs.+ - [Remove the background around hair online](https://www.cutoutlabs.com/remove-background-from-hair): Remove portrait backgrounds without losing fine hair: inspect soft transparency and troubleshoot missing strands, halos, and background color spill.+ - [Client-side mask upscaling for 16 MP cutouts](https://www.cutoutlabs.com/high-resolution-background-removal): Learn how a smaller upload proxy and local compositing support cutouts up to 16 MP, and why full-size exports still need edge inspection.+ - [Remove the background from car photos for vehicle listings](https://www.cutoutlabs.com/remove-background-from-car-photos): Cut a car out of a street or driveway photo, decide what to do with windows and the ground shadow, and export a clean listing image on white, a color, or a new scene.+ - [Remove the background from clothing photos](https://www.cutoutlabs.com/remove-background-from-clothing-photos): Cut jackets, shirts, and dresses out of hanger, mannequin, or flat-lay photos, remove hands and hangers, and catch the wall that hides between a sleeve and the body.+ - [Remove the background from pet photos](https://www.cutoutlabs.com/remove-background-from-pet-photos): Cut a dog, cat, or other pet out of a photo while keeping fur, ears, and whiskers, then place it on a color or new scene for portraits, cards, and prints.+ - [Background removal MCP server for Claude, Cursor, and VS Code](https://www.cutoutlabs.com/background-removal-mcp-server): Connect the Cutout Labs MCP server so an AI assistant can remove image backgrounds from URLs or local files. Setup for Claude Desktop, Cursor, and VS Code, with tool options and limits.+ - [Background remover vs. AI image generators: which to use for product cutouts](https://www.cutoutlabs.com/background-remover-vs-ai-image-generators): A segmentation tool hides background pixels and keeps the product's own; a generative image editor redraws the picture and can change logos, labels, and small text. How to tell which you need and how to check.+ ## Blog+ - [Can Gemini or GPT image models replace a background remover? We tested 50 photos](https://www.cutoutlabs.com/blog/ai-image-generator-vs-background-removal-benchmark): We ran 50 product and lifestyle photos through Cutout Labs, two Gemini image models, and two GPT image models. The AI image models were 2–7× slower, cost $39–68 per 1,000 images, and the ones that returned real transparency did it by redrawing the product.+ - [Where automatic background removal fails, and how to fix it: 33 tested requests](https://www.cutoutlabs.com/blog/background-removal-limitations-tested): We ran 23 photos through the Cutout Labs API to find where automatic cutouts go wrong: hands, competing subjects, plated food, cropped furniture, glass, mesh, motion blur, and low resolution. Each case shows the real result and a fix.+ - [The Ghost Hand Problem: Why AI Background Removers Strip Hands from Product Photos](https://www.cutoutlabs.com/blog/why-ai-background-removers-strip-hands): Why background removal can omit the hand holding a product, how foreground ambiguity contributes, and what developers can try.+ - **Which formats can I download?** PNG keeps transparency. JPG and WebP are smaller; JPG always uses a solid background. PDF suits print and documents and keeps transparency. You can also copy the cutout straight to your clipboard.
  • server.json 2 fields changed
    repository_url
    — https://github.com/cutout-labs/mcp
    version
    0.2.1 0.2.2