PromptShopMCP

Use AI to edit image in Claude Desktop / Cursor (AI P图)

python

18

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "kira-pgr-promptshopmcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/project/dir/",
        "run",
        "mcp",
        "run",
        "/project/dir/server.py"
      ],
      "env": {
        "GEMINI_API_KEY": "key",
        "REMOVEBG_API_KEY": "key",
        "FREEIMAGE_API_KEY": "key"
      }
    }
  }
}

PromptShopMCP is an MCP server that converts text prompts into image transformations. You can generate new images, modify existing ones, remove backgrounds, and host/share results using a streamlined, command-driven workflow. This server is designed to be integrated with MCP clients so you can control image tasks with natural language instructions.

How to use

Connect to the PromptShopMCP server from an MCP client. You have two practical options for running the server locally or from your tooling: a standard stdio setup via UV and a local script-based integration via Cursor.

Standard stdio setup with UV. Run the MCP server directly from a project directory using UV. You provide the directory path, start the MCP, and point to the server script. You must supply API keys for Gemini, remove.bg, and FreeImage.host as environment variables when starting the server.
Cursor-based integration. Place a script in your project and configure Cursor to run the MCP server. You will set the absolute path to the script and ensure the API keys and project directory are available to the server at startup.

How to install

Prerequisites: Python 3.11 or higher is required.

# Prerequisites
python3.11 --version
# Ensure you have Python 3.11+ installed on your system

Install UV and set up a Python virtual environment for the MCP server.

# On macOS and Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# On Windows
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

# With pip (alternative to manage UV)
pip install uv

Create and activate a virtual environment, then install dependencies for the server.

uv venv --python=python3.11
source .venv/bin/activate  # or .venv\Scripts\activate on Windows
uv pip install -r requirements.txt

Configuration and runtime examples

You can run the MCP server in a couple of ways. The following options are explicit configuration examples you can adapt for your setup.

{
  "PromptShopMCP": {
    "command": "uv",
    "args": [
      "--directory",
      "/project/dir/",
      "run",
      "mcp",
      "run",
      "/project/dir/server.py"
    ],
    "env": {
      "GEMINI_API_KEY": "key",
      "FREEIMAGE_API_KEY": "key",
      "REMOVEBG_API_KEY": "key"
    }
  }
}

Cursor-based setup uses a local script. The command to start the server is the shell to run the script, and you must provide the absolute path to the script.

# Linux/macOS example
sh /absolute/path/to/cursor.sh

# Windows example
cmd /c C:\absolute\path\to\cursor.bat

Available tools

generate_image

Create new images from text prompts using Gemini models.

modify_image

Transform existing images according to text instructions to achieve the desired edit.

remove_background

Remove backgrounds from images using the remove.bg API.

host_image

Share generated images via FreeImage.host.

manage_resources

Track and manage generated and uploaded images to keep an overview of your assets.