Easy Email Builder
LocalAbout Easy Email Builder
Easy Email Builder is an MCP server for AI assistants that create HTML email templates. It validates and renders templates, returning HTML, a visual-editor preview link, variables, and optionally editor JSON.
Key features
- 01Reference template blocks and properties
- 02Browse 31 starter templates
- 03Validate and render email HTML
- 04Return template variables and fallbacks
- 05Create preview links for visual editing
Use cases
- 01Draft a welcome email from a prompt
- 02Check email templates for client compatibility issues
- 03Open a generated template in the visual editor
MCP support
The server exposes tools for template references, listing and retrieving templates, and building emails. MCP clients can connect using a local server setup.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y easy-email-builder-mcpEASY_EMAIL_BUILDER_EDITOR_URLEditor URL for preview links, if you host the editor yourself. Defaults to the public demo.Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
4Saved tool list · updated 2w ago
This tool takes no parameters.
Details
From the project READMEEasy Email Builder MCP server
Let AI assistants (Claude, Cursor, VS Code, …) design HTML email templates with Easy Email Builder. The assistant writes the template, the server validates and renders it, and you get:
- the final email HTML,
- a preview link that opens the email in the visual editor, where you can keep editing and export it,
- the template's variables (
{{first_name}},{%unsubscribe_link%}, …) with their fallback values, - optionally the editor JSON, to load into the
EmailBuilderReact component.
Add it to your AI agent
Works with any agent that supports local MCP servers. The command-line setups need Node.js 18+.
Claude Desktop: one-click extension (easiest)
- Download easy-email-builder.mcpb.
- Double-click it (or drag it into Claude Desktop → Settings → Extensions) and click Install.
No Node.js or config editing needed: Claude Desktop runs it with its built-in Node.
Claude Code
claude mcp add easy-email-builder --scope user -- npx -y easy-email-builder-mcp
--scope user makes it available in all your projects; leave it out to add it to the current project only.
OpenAI Codex
codex mcp add easy-email-builder -- npx -y easy-email-builder-mcp
or in ~/.codex/config.toml:
[mcp_servers.easy-email-builder]
command = "npx"
args = ["-y", "easy-email-builder-mcp"]
VS Code (GitHub Copilot)
code --add-mcp '{"name":"easy-email-builder","command":"npx","args":["-y","easy-email-builder-mcp"]}'
or .vscode/mcp.json in your project:
{
"servers": {
"easy-email-builder": {
"command": "npx",
"args": ["-y", "easy-email-builder-mcp"]
}
}
}
Gemini CLI
gemini mcp add easy-email-builder npx -y easy-email-builder-mcp
Cursor, Windsurf, Cline and others
Add this to the client's MCP config:
{
"mcpServers": {
"easy-email-builder": {
"command": "npx",
"args": ["-y", "easy-email-builder-mcp"]
}
}
}
| Client | Config file |
|---|---|
| Cursor | ~/.cursor/mcp.json (or .cursor/mcp.json in a project) |
| Windsurf | ~/.codeium/windsurf/mcp_config.json |
| Cline / Roo Code | MCP Servers panel → Configure |
| Claude Desktop (manual) | Settings → Developer → Edit Config |
If the server fails to start with spawn npx ENOENT or env: node: No such file or directory (common with nvm, because desktop apps don't load your shell's PATH), use the full path from which npx as the command and add your Node folder to PATH:
{
"mcpServers": {
"easy-email-builder": {
"command": "/Users/you/.nvm/versions/node/v22.0.0/bin/npx",
"args": ["-y", "easy-email-builder-mcp"],
"env": { "PATH": "/Users/you/.nvm/versions/node/v22.0.0/bin:/usr/local/bin:/usr/bin:/bin" }
}
}
}
ChatGPT and claude.ai (web): these only connect to MCP servers hosted at a URL. This server runs on your machine, so use it from Claude Desktop, Claude Code, Codex or one of the clients above.
Try it
Ask, for example: "Create a welcome email for my coffee shop with a 10% discount code and the customer's first name."
Tools
| Tool | What it does |
|---|---|
get_block_reference | The template format: blocks, properties, colors, variables |
list_templates | 31 starter templates, plus the built-in header and footer ids |
get_template | A starter template to adapt (nested "tree" format or flat editor JSON) |
build_email | Validates a template, renders the HTML, lists its variables and returns a preview link. Warns about things that break in Outlook or Gmail (nested columns, oversized images, unsupported CSS, Gmail clipping). Can also save the HTML/JSON to files |
Preview links
The template is compressed into the link itself (…/Easy-Email-Builder/#z/…). Nothing is uploaded or stored on a server. Very large templates make long links, and some apps cut those off; build_email warns when a link is long.
To open links in your own deployment of the editor, set EASY_EMAIL_BUILDER_EDITOR_URL:
"env": { "EASY_EMAIL_BUILDER_EDITOR_URL": "https://email.example.com/editor/" }
Your deployment must decode #z/ links; see decodeTemplateHash in src/getConfiguration.
Development
The server reuses the editor's own code from ../src (rendering, templates, variables) and bundles it into a single file.
pnpm install # in the repo root (email block packages)
cd mcp && pnpm install
pnpm test # build, pack the extension, end-to-end tests over stdio
pnpm pack:extension # build dist/easy-email-builder.mcpb (Claude Desktop extension)
pnpm typecheck
Try it with the MCP Inspector: npx @modelcontextprotocol/inspector node dist/index.js.
Releasing
- Set the same version in
package.jsonand inserver.json(versionandpackages[0].version). The extension takes its version frompackage.json. pnpm test, thennpm publish --otp=<code>.mcp-publisher publish(after npm: the registry checks the npm package).- Create a GitHub release and attach
dist/easy-email-builder.mcpb(keep that file name: the README's download link points to the latest release'seasy-email-builder.mcpb).
License
MIT
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.