What can you do with this server?

This server lets AI assistants remove image backgrounds and check Poof account credits. * remove_background: Remove background from an image supplied as a base64 string or URL. * Choose output format: png, jpg, or webp. * Output rgba (transparent) or rgb (opaque, with optional bg_color). * Resize via size presets (full, preview, medium, hd) or exact width/ height. * Fit behavior with fit: contain, cover, or scale-down. * Optionally crop to subject bounds. * get_account: Show account plan, credit usage, remaining credits, and auto-recharge threshold. * Supports local stdio and remote HTTP/Cloudflare Worker transports.

How do I use Poof Background Removal MCP Server?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Poof Background Removal MCP Server remove the background from https://example.com/portrait.jpg" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

Poof Background Removal MCP Server

by poof-bg

Overview Schema Source Related Servers Discussions Admin

TypeScript

Hybrid

@poof-bg/mcp

MCP (Model Context Protocol) server for the Poof background removal API. Use AI assistants like Claude to remove backgrounds from images.

Migrating from remove.bg? It shuts down on 1 December 2026 — see the remove.bg alternative and migration guide.

Features

remove_background - Remove background from images (accepts base64 or URL)
get_account - Check your account info and credit balance
Supports both stdio (local) and HTTP (remote/Cloudflare Worker) transports

Related MCP server: simplypng-mcp

Installation

Local (stdio transport)

npm install -g @poof-bg/mcp

Or install locally:

git clone https://github.com/poof-bg/mcp.git
cd mcp
npm install
npm run build

Remote (HTTP transport via Cloudflare Worker)

The Poof MCP server can be deployed as a Cloudflare Worker, allowing remote access via HTTP.

Configuration

Environment Variable

Set your Poof API key:

export POOF_API_KEY=your_api_key_here

Get your API key at dash.poof.bg

Claude Code

OAuth (recommended)

claude mcp add --transport http poof https://api.poof.bg/mcp

API Key

claude mcp add --transport http poof https://api.poof.bg/mcp \
  --header "x-api-token: YOUR_POOF_API_KEY"

Claude Desktop

Add to your Claude Desktop config file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

Option 1: Remote with OAuth (recommended)

{
  "mcpServers": {
    "poof": {
      "url": "https://api.poof.bg/mcp"
    }
  }
}

Option 2: Local (stdio transport)

{
  "mcpServers": {
    "poof": {
      "command": "npx",
      "args": ["-y", "@poof-bg/mcp"],
      "env": {
        "POOF_API_KEY": "your_api_key_here"
      }
    }
  }
}

Option 3: Remote with API key

{
  "mcpServers": {
    "poof": {
      "url": "https://api.poof.bg/mcp",
      "headers": {
        "x-api-token": "YOUR_POOF_API_KEY"
      }
    }
  }
}

Cursor

Add to .cursor/mcp.json in your project root (or global config):

OAuth (recommended)

{
  "mcpServers": {
    "poof": {
      "url": "https://api.poof.bg/mcp"
    }
  }
}

Local (stdio)

{
  "mcpServers": {
    "poof": {
      "command": "npx",
      "args": ["-y", "@poof-bg/mcp"],
      "env": {
        "POOF_API_KEY": "your_api_key_here"
      }
    }
  }
}

Remote with API key

{
  "mcpServers": {
    "poof": {
      "url": "https://api.poof.bg/mcp",
      "headers": {
        "x-api-token": "YOUR_POOF_API_KEY"
      }
    }
  }
}

Windsurf

Add to ~/.codeium/windsurf/mcp_config.json:

OAuth (recommended)

{
  "mcpServers": {
    "poof": {
      "serverUrl": "https://api.poof.bg/mcp"
    }
  }
}

Local (stdio)

{
  "mcpServers": {
    "poof": {
      "command": "npx",
      "args": ["-y", "@poof-bg/mcp"],
      "env": {
        "POOF_API_KEY": "your_api_key_here"
      }
    }
  }
}

Remote with API key

{
  "mcpServers": {
    "poof": {
      "serverUrl": "https://api.poof.bg/mcp",
      "headers": {
        "x-api-token": "YOUR_POOF_API_KEY"
      }
    }
  }
}

VS Code + Copilot

Add to your VS Code settings.json:

OAuth (recommended)

{
  "mcp": {
    "servers": {
      "poof": {
        "url": "https://api.poof.bg/mcp"
      }
    }
  }
}

Local (stdio)

{
  "mcp": {
    "servers": {
      "poof": {
        "command": "npx",
        "args": ["-y", "@poof-bg/mcp"],
        "env": {
          "POOF_API_KEY": "your_api_key_here"
        }
      }
    }
  }
}

Remote with API key

{
  "mcp": {
    "servers": {
      "poof": {
        "url": "https://api.poof.bg/mcp",
        "headers": {
          "x-api-token": "YOUR_POOF_API_KEY"
        }
      }
    }
  }
}

Cline (VS Code Extension)

Open Cline settings and add to the MCP Servers configuration:

OAuth (recommended)

{
  "poof": {
    "url": "https://api.poof.bg/mcp"
  }
}

Local (stdio)

{
  "poof": {
    "command": "npx",
    "args": ["-y", "@poof-bg/mcp"],
    "env": {
      "POOF_API_KEY": "your_api_key_here"
    }
  }
}

Remote with API key

{
  "poof": {
    "url": "https://api.poof.bg/mcp",
    "headers": {
      "x-api-token": "YOUR_POOF_API_KEY"
    }
  }
}

Tools

remove_background

Remove the background from an image.

Parameters:

Parameter	Type	Required	Description
`image`	string	Yes	Base64-encoded image or URL to an image
`format`	string	No	Output format: `png`, `jpg`, `webp` (default: `png`)
`channels`	string	No	Color channels: `rgba` for transparency, `rgb` for opaque (default: `rgba`)
`bg_color`	string	No	Background color when using `rgb` channels (e.g., `#ffffff`)
`size`	string	No	Output size preset: `full`, `preview`, `medium`, `hd` (default: `full`). Ignored when `width`/`height` is set
`crop`	boolean	No	Crop to subject bounds (default: `false`)
`width`	integer	No	Output width in pixels (1-6000). Alone, the height follows the aspect ratio
`height`	integer	No	Output height in pixels (1-6000). Alone, the width follows the aspect ratio
`fit`	string	No	How to fit into `width` x `height` without stretching: `contain` (default, pad), `cover` (fill, crop the overflow around the subject), `scale-down` (pad, never enlarge)

Example prompts:

Remove the background from this image: https://example.com/photo.jpg

Remove the background and add a white background instead

Remove the background and give me a 500x500 image with the product centred

get_account

Get your account information and credit balance.

Example prompts:

Check my Poof account balance

How many credits do I have left?

Response:

{
  "success": true,
  "data": {
    "organizationId": "org_abc123",
    "plan": "Pro",
    "maxCredits": 5000,
    "usedCredits": 1234,
    "remainingCredits": 3766,
    "autoRechargeThreshold": 100
  }
}

Development

# Install dependencies
npm install

# Build
npm run build

# Run in dev mode (Cloudflare Worker)
npm run dev

# Run locally (stdio mode)
POOF_API_KEY=your_key npm start

# Lint
npm run lint

# Format
npm run format

Deployment (Cloudflare Worker)

To deploy the MCP server as a Cloudflare Worker:

# Install Wrangler if you haven't already
npm install -g wrangler

# Login to Cloudflare
wrangler login

# Deploy
wrangler deploy

Set the required environment secrets:

wrangler secret put POOF_API_KEY
wrangler secret put MCP_JWT_SECRET
wrangler secret put COOKIE_ENCRYPTION_KEY

Also set the dashboard URL variable:

wrangler secret put DASHBOARD_URL

You'll also need to create a KV namespace for OAuth state storage and update the IDs in wrangler.toml:

wrangler kv namespace create OAUTH_KV
wrangler kv namespace create OAUTH_KV --preview

License

MIT

Available Tools

2 tools

get_accountB

Get account information including plan details and credit usage

ParametersJSON Schema

Name	Required	Description	Default
No parameters

TDQS

B3.2/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden of behavioral disclosure. While 'Get' implies a read operation, the description doesn't specify whether this requires authentication, has rate limits, returns real-time or cached data, or what format the information comes in. For a tool with zero annotation coverage, this represents significant gaps in behavioral transparency.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is perfectly concise - a single sentence that efficiently communicates the tool's purpose without any wasted words. It's front-loaded with the core functionality and includes specific details about what information is retrieved.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a simple read operation with no parameters, the description covers the basic purpose adequately. However, with no annotations and no output schema, the description doesn't address important behavioral aspects like authentication requirements, response format, or data freshness. The description is complete enough for the tool's simplicity but leaves gaps in operational context.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The tool has zero parameters, and schema description coverage is 100% (though trivial since there are no parameters). The description appropriately doesn't discuss parameters since none exist. A baseline of 4 is appropriate for a zero-parameter tool where the schema fully documents the parameter situation.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose with a specific verb ('Get') and resource ('account information'), including what information is retrieved ('plan details and credit usage'). It doesn't differentiate from its only sibling 'remove_background', but that's reasonable since they serve completely different functions.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides no guidance on when to use this tool versus alternatives. While there's only one sibling tool ('remove_background') that serves a completely different purpose, the description doesn't mention any prerequisites, constraints, or contextual factors for using this tool.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

remove_backgroundA

Remove the background from an image. Returns the processed image as base64. Accepts URL or base64-encoded image. Can also resize the result to an exact width x height without stretching (width, height, fit).

ParametersJSON Schema

Name	Required	Description	Default
`fit`	No	How to fit the image into width x height. 'contain' pads the remaining area (transparent or bg_color), 'cover' fills the canvas and crops the overflow around the subject, 'scale-down' is like contain but never enlarges a smaller image	contain
`crop`	No	Whether to crop the image to the subject bounds
`size`	No	Output image size preset (megapixel cap). Ignored when width or height is set	full
`image`	Yes	Image input: base64-encoded image data or a URL to an image
`width`	No	Output width in pixels. On its own, the height follows the aspect ratio. With height, the image is fitted into an exact width x height canvas without stretching
`format`	No	Output image format	png
`height`	No	Output height in pixels. On its own, the width follows the aspect ratio. With width, the image is fitted into an exact width x height canvas without stretching
`bg_color`	No	Background color (hex, rgb, or color name). Only applies when channels is 'rgb'. Example: '#ffffff'
`channels`	No	Output color channels. Use 'rgba' for transparency, 'rgb' for opaque background	rgba

TDQS

A4/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden. It does disclose key behavioral traits: 'Returns the processed image as base64' and 'Accepts URL or base64-encoded image.' However, it does not mention potential error conditions, limits, side effects, or whether the operation is stateless. It adds some behavioral transparency but not comprehensive context.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is two sentences with no filler. The primary purpose is front-loaded, followed by output format, input types, and optional resizing capability. Every sentence contributes useful information, and the structure makes it easy to scan.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with 9 parameters and no output schema, the description provides a complete high-level view: what it does, accepted inputs, return format, and the key resizing behavior. The schema supplies detailed parameter semantics, so the description need not repeat them. It lacks explicit error handling or limitations, but the essential information for correct invocation is present.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the baseline is 3. The description mentions width, height, and fit in passing but does not add meaning beyond the schema's detailed parameter descriptions. The schema already fully documents each parameter, so the description adds minimal additional semantic value.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses a specific verb and resource: 'Remove the background from an image.' It clearly distinguishes itself from the only sibling, get_account, which serves a completely different purpose. The statement is unambiguous and instantly conveys the tool's core function.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides clear context about when to use the tool: whenever background removal is needed. It states the accepted input types (URL or base64) and the available resizing behavior, giving enough context for an agent to select it. It does not explicitly mention when not to use it or list alternatives, but no close alternatives exist among siblings, so this is a minor gap.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

1 tool updatev1.2.0Sep 8, 2026

Changedremove_background5 fields changed

addedInput schema / properties / fit

Added value: +{
+  "default": "contain",
+  "description": "How to fit the image into width x height. 'contain' pads the remaining area (transparent or bg_color), 'cover' fills the canvas and crops the overflow around the subject, 'scale-down' is like contain but never enlarges a smaller image",
+  "enum": [
+    "contain",
+    "cover",
+    "scale-down"
+  ],
+  "type": "string"
+}

addedInput schema / properties / height

Added value: +{
+  "description": "Output height in pixels. On its own, the width follows the aspect ratio. With width, the image is fitted into an exact width x height canvas without stretching",
+  "maximum": 6000,
+  "minimum": 1,
+  "type": "integer"
+}

changedInput schema / properties / size / description

Previous value: -"Output image size preset"New value: +"Output image size preset (megapixel cap). Ignored when width or height is set"

changedInput schema / properties / size / enum

Previous value: -[
-  "full",
-  "preview",
-  "small",
-  "medium",
-  "large"
-]New value: +[
+  "full",
+  "preview",
+  "medium",
+  "hd"
+]

addedInput schema / properties / width

Added value: +{
+  "description": "Output width in pixels. On its own, the height follows the aspect ratio. With height, the image is fitted into an exact width x height canvas without stretching",
+  "maximum": 6000,
+  "minimum": 1,
+  "type": "integer"
+}

2 tool updatesv1.1.0Feb 19, 2026
- First observedget_account
- First observedremove_background

Deploy Server

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.7/5.0

Scored 2026-09-08 15:50 across 2 tools

Disambiguation5/5

The two tools have clearly distinct purposes: one handles account/plan information and the other performs the core background removal operation. There is no overlap or ambiguity between them.

Naming Consistency5/5

Both tools follow a consistent verb_noun snake_case pattern: get_account and remove_background. The naming is predictable and immediately conveys each tool's action and target.

Tool Count3/5

With only two tools, the server feels minimal but appropriate for a focused single-purpose service. The account tool supports the main workflow, yet the surface is borderline thin compared to the typical well-scoped range.

Completeness4/5

The core background removal workflow is fully covered, including input via URL or base64 and optional resizing. Account and credit information is also available, though there are minor missing conveniences such as batch processing or output format options.

Maintenance

ActivityMaintained

ResponsivenessNo issues

Resources

Need Help?

Related Servers

Related MCP Connectors

Maskr.io MCP Server AOAuth
io.maskr.mcp
AI-powered image processing via GPU. Remove backgrounds and upscale images (2x/4x) directly from any MCP client. OAuth 2.1 authenticated, returns processed images inline with download links. Free credits on signup at maskr.io.
Image & Video Processing Multimedia Processing App Automation
62
apiai.meOAuth
me.apiai
Image and video AI tools and your own pipelines, run from any AI assistant.
Image & Video Processing AI & Machine Learning
1
SammaPix — Image & PDF tools for AI agents
io.github.samma1997
Image & PDF tools for AI agents: compress, convert, resize, PDF, AI vision, pipeline.
Image & Video Processing AI & Machine Learning
17
Remover Fundo A
io.github.mcp-dir
Removes the background from an image by URL, cutting out the foreground person or object with AI and
Image & Video Processing AI & Machine Learning
71

Related MCP Servers

Rembg MCP Server
Image & Video Processing Multimedia Processing App Automation
holocode-ai
A
license
B
quality
D
maintenance
Enables AI-powered background removal from images using multiple specialized models including u2net, birefnet, and isnet. Supports both single image processing and batch folder operations with advanced options like alpha matting and mask-only output.
Updated a year ago (2025-09-18 11:00 UTC)
2
4
MIT
simplypng-mcpofficial
Image & Video Processing AI & Machine Learning
SimplyPNG
A
license
A
quality
C
maintenance
MCP server for SimplyPNG background removal APIs. Enables AI agents to remove backgrounds from images, estimate credits, check job status, and download results programmatically.
Updated 15 days ago (2026-09-13 13:17 UTC)
5
17 npm
MIT
Recraft MCP Server
Image & Video Processing AI & Machine Learning
runapi-ai
A
license
A
quality
B
maintenance
Enables background removal and upscaling of images using Recraft models via RunAPI, with task polling and pricing lookup.
Updated 2 months ago (2026-07-31 09:38 UTC)
5
62 npm
Apache 2.0
magic-wand
Image & Video Processing Design Tools
nathandstory
A
license
Not graded
quality
D
maintenance
Enables AI agents to interactively remove image backgrounds using a Paint.NET-style magic wand tool with live visual feedback and precision helpers.
Updated 3 months ago (2026-07-10 19:52 UTC)
MIT

@poof-bg/mcp

Features

Installation

Local (stdio transport)

Remote (HTTP transport via Cloudflare Worker)

Configuration

Environment Variable

Claude Code

OAuth (recommended)

API Key

Claude Desktop

Option 1: Remote with OAuth (recommended)

Option 2: Local (stdio transport)

Option 3: Remote with API key

Cursor

OAuth (recommended)

Local (stdio)

Remote with API key

Windsurf

OAuth (recommended)

Local (stdio)

Remote with API key

VS Code + Copilot

OAuth (recommended)

Local (stdio)

Remote with API key

Cline (VS Code Extension)

OAuth (recommended)

Local (stdio)

Remote with API key

Tools

remove_background

get_account

Development

Deployment (Cloudflare Worker)

License

Links

Available Tools

TDQS

TDQS

Tool Schema Changelog

TDQS

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

Rembg MCP Server

simplypng-mcpofficial

Recraft MCP Server

magic-wand