AI-powered image asset generation MCP server with prompt engineering guidance and vectorization
npm install @morphkit/asset-gen-mcpAI-powered image asset generation with prompt engineering guidance and vectorization
Run the MCP server directly without global installation:
``bash`
npx -y @morphkit/asset-gen-mcp@latest
For MCP client configuration, use:
`json`
{
"mcpServers": {
"asset-gen": {
"command": "npx",
"args": ["-y", "@morphkit/asset-gen-mcp@latest"],
"env": {
"GOOGLE_API_KEY": "YOUR_API_KEY"
}
}
}
}
For Claude Code:
`bash`
claude mcp add asset-gen -e GOOGLE_API_KEY=YOUR_API_KEY -- npx -y @morphkit/asset-gen-mcp@latest
- get-image-generation-prompt-instructions: Returns asset-type-specific prompt engineering best practices and examples for Google Imagen
- generate-images: Generates images using Google Imagen 4.0 with auto-vectorization for SVG outputs
- vectorize-image: Converts raster images to SVG using configurable vectorization options
- remove-background: AI-powered background removal using transparent-background
- Node.js >= 18
- Python (for background removal) - Pre-installed on macOS and most Linux distributions
- Google API Key - Get one at Google AI Studio
On first use, the remove-background tool downloads an ML model (~150MB) and creates a Python virtual environment. This takes approximately 3 minutes. Subsequent runs are fast (~2-5 seconds).
\\\bash
cd packages/asset-gen-mcp
bun install
\\\
Copy the example environment file and configure:
\\\bash
cp .env.example .env
\\\
\\\bash
bun run dev
\\\
\\\bash
bun run build
\\\
\\\bash
bun run start
\\\
Returns asset-type-specific prompt engineering best practices and examples for Google Imagen
Usage:
\\\json
{
"name": "get-image-generation-prompt-instructions",
"arguments": {
"param": "value"
}
}
\\\
Generates images using Google Imagen 4.0 with auto-vectorization for SVG outputs
Usage:
\\\json
{
"name": "generate-images",
"arguments": {
"param": "value"
}
}
\\\
Converts raster images to SVG using configurable vectorization options
Usage:
\\\json
{
"name": "vectorize-image",
"arguments": {
"param": "value"
}
}
\\\
AI-powered background removal using InSPyReNet model. Works with complex backgrounds, shadows, and lighting effects. Outputs transparent PNG.
Usage:
\\\json
{
"name": "remove-background",
"arguments": {
"inputPath": "/path/to/image.png",
"outputPath": "/path/to/output.png",
"fast": false
}
}
\\\
Set fast: true for quicker processing with slightly lower quality (384x384 model vs 1024x1024).
\\\bash
bun run test
\\\
\\\bash
bun run check-types
\\\
\\\bash
bun run lint
\\\`