About SVG.new
SVG.new is an MCP server that provides image vectorization and SVG editing for AI agents and design workflows. Image and SVG files can be processed through its hosted endpoint or a local server.
Key features
- 01Convert PNG, JPG, and WebP to SVG
- 02Recolor SVGs and simplify palettes
- 03Remove image backgrounds
- 04Batch-vectorize images
- 05Check batch status and history
Use cases
- 01Convert raster images into scalable SVGs
- 02Adjust SVG colors or reduce palettes
- 03Process image batches and track progress
MCP support
The server exposes MCP tools for image vectorization, SVG recoloring and simplification, background removal, and batch processing. Clients can connect to its hosted Streamable HTTP endpoint or run the local server.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://svg.new/api/mcpclaude mcp add --transport http svgnew https://svg.new/api/mcp{"mcpServers":{"svgnew":{"type":"http","url":"https://svg.new/api/mcp"}}}Open in Cursor{"servers":{"svgnew":{"type":"http","url":"https://svg.new/api/mcp"}}}Open in VS Code{"mcpServers":{"svgnew":{"type":"http","url":"https://svg.new/api/mcp"}}}{"mcpServers":{"svgnew":{"serverUrl":"https://svg.new/api/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project README@svgnew/mcp
MCP (Model Context Protocol) server for svg.new — AI-powered image to SVG vectorization.
Use svg.new from any MCP-compatible client: Claude Code, Claude Desktop, Cursor, Windsurf, and more.
Hosted server (sign in, no API key)
SVG.new hosts this server at https://svg.new/api/mcp (Streamable HTTP). Clients that support OAuth sign in with your SVG.new account; there is nothing to install. Clients that send headers can use Authorization: Bearer svk_... with an API key instead.
# Claude Code (then run /mcp, pick svgnew and choose Authenticate)
claude mcp add --transport http --scope user svgnew https://svg.new/api/mcp
# Gemini CLI (then run /mcp auth svg-new)
gemini extensions install https://github.com/svgnew/mcp
In Claude, open Settings, Connectors, Add custom connector, enter the URL above and sign in. Cline and other agents can follow llms-install.md. The cursor folder is the Cursor plugin for the hosted server. Setup for each client: svg.new/mcp.
Quick Start (local server)
npx @svgnew/mcp
Requires SVG_NEW_API_KEY environment variable. Get your key at svg.new/account.
Setup
Claude Code / Claude Desktop
Add to your MCP config:
{
"mcpServers": {
"svg-new": {
"command": "npx",
"args": ["@svgnew/mcp"],
"env": {
"SVG_NEW_API_KEY": "svk_your_key_here"
}
}
}
}
Cursor / Windsurf
Add to your MCP settings with the same configuration above.
OpenAI Codex CLI
Add svg.new to Codex:
codex mcp add svg-new -- npx -y @svgnew/mcp
Or install as a plugin:
codex install svg-new
Set your API key:
codex mcp add svg-new --env SVG_NEW_API_KEY=svk_your_key_here -- npx -y @svgnew/mcp
Available Tools
| Tool | Description | Cost |
|---|---|---|
vectorize | Convert image to SVG | 1 credit |
vectorize_file | Convert a local image file to SVG | 1 credit |
recolor_svg | Change colors in SVG | Free |
simplify_svg | Reduce color palette | Free |
remove_background | AI background removal | 1 credit |
remove_background_file | Background removal from a local image file | 1 credit |
batch_vectorize | Process multiple images | 1 credit/image |
batch_vectorize_files | Vectorize up to 10 local image files | 1 credit/image |
get_batch_status | Poll batch progress | Free |
list_vectorizations | View history | Free |
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
SVG_NEW_API_KEY | Yes | — | Your svg.new API key |
SVG_NEW_API_URL | No | https://svg.new | API base URL |
Links
- svg.new — Web app
- Pricing — Plans and credits
- Claude Code Plugin — Full plugin with skills
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.