# lassocut-mcp [Health: Active]

**Category:** 💻 Developer Tools  
**Repository:** https://github.com/jaupin94/lassocut-mcp  
**GitHub Stars:** 0  
**npm Downloads (last month):** 450  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/lassocut-mcp

## Description
Remove image backgrounds from AI assistants with a LassoCut account (files, folders, URLs).

## Tools
Capabilities this server exposes over MCP:

- **remove_background** — Remove the background of one or more images.
- **get_credits** — Shows your current credit balance and the free previews available now (50 a month, up to 10 a day).
- **sign_in** — Connects a LassoCut account: opens the approval page in your browser and returns immediately with the link and code so you can approve at your own pace. Pass `force: true` to switch to a different account.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "lassocut-mcp": {
    "command": "npx",
    "args": ["-y","lassocut-mcp"],
    "env": {
      "LASSOCUT_API_KEY": ""
    }
  }
}
```

**Requires environment variables:** `LASSOCUT_API_KEY` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What lassocut-mcp MCP server does

The lassocut-mcp MCP server lets an AI assistant send images to LassoCut for background removal. The `remove_background` tool accepts individual file paths, folders, HTTP or HTTPS image URLs, or a list containing these input types. A call can process up to 50 images, while files larger than 22 MB are skipped.

Results are written as new files rather than replacing the originals. For local inputs, output is saved beside each source unless an absolute `output_dir` is provided. URL-based images are saved to the Downloads folder by default, with the home folder used as a fallback. Generated names use the `-no-bg` suffix and add `-2`, `-3`, and so on when a filename already exists.

## How it works

The server communicates with the LassoCut API, which is compatible with the remove.bg API. Images are sent only to LassoCut for processing. The default output is a low-cost preview, using 0.25 credit per image. Full-size output costs 1 credit per image and is selected only when requested.

`remove_background` supports transparent output by default, a named or hexadecimal background color, and PNG, JPG, or WebP output. The `crop` option can trim the result around the detected subject. Up to three small thumbnails may be returned inline with a result.

Full-size requests covering more than 10 images require explicit cost confirmation through `confirm_cost`. The server also checks whether the account has enough credits before sending a full-size request. Preview availability is limited to 50 free previews per month and up to 10 per day.

## Setup and configuration

Install the lassocut-mcp MCP server with the downloadable Claude Desktop package, the `claude mcp add` command for Claude Code, or an MCP JSON configuration using `npx -y lassocut-mcp`. Node.js 20 or newer is required. Cursor and Windsurf can use the same command-based JSON configuration.

A LassoCut account and API key are required. The `sign_in` tool opens an approval page in the browser and returns a link and short code immediately. After approval, the key is saved automatically. Passing `force: true` starts account switching. Existing credentials from the `lassocut login` command are also recognized. Setting `LASSOCUT_API_KEY`, or configuring `api_key` in Claude Desktop extension settings, takes precedence over a saved login.

## Tools and capabilities

- `remove_background`: Processes one or more local files, folders, or URLs; controls size, background, format, cropping, and output location.
- `get_credits`: Reports the current credit balance and available free previews.
- `sign_in`: Starts browser-based LassoCut authorization and can switch accounts.

## Limitations and notes

The lassocut-mcp MCP server does not process subfolders when a folder path is supplied. Files above 22 MB are skipped, and full-size jobs over 10 images need user approval after the cost is displayed. Images are sent only to the LassoCut API; users should consult LassoCut's privacy policy for service-specific data handling details.

The lassocut-mcp MCP server uses previews by default, so assistants should request full size only when the higher resolution is needed. Existing files are protected from overwrite, but output naming can create numbered variants when matching names are already present.

_Full upstream README: https://allmcps.com/mcp/lassocut-mcp/readme_