Manage Translations from Your AI Coding Assistant with Verba's MCP Server

Last updated: February 2026

If you use Claude Code, Cursor, or any MCP-compatible AI assistant, you can now manage your Verba (verba.dev) translations without leaving your editor.

No browser tabs. No dashboard. Just tell your AI assistant what you need.

What is MCP?

Model Context Protocol (MCP) is an open standard that lets AI assistants connect to external tools. It was created by Anthropic and is now maintained under the Linux Foundation. Claude Desktop, Claude Code, Cursor, Windsurf, and other AI tools all support MCP.

An MCP server exposes "tools" — functions the AI can call on your behalf. The Verba MCP server exposes 9 tools for managing translations.

What You Can Do

Once connected, you can ask your AI assistant things like:

"Show me all untranslated keys for French" — lists every key missing a French translation
"Add a key 'checkout.success' with value 'Payment successful!' and translate it" — creates the key and triggers AI translation to all your languages
"What languages does my project support? Add Japanese." — lists locales and adds a new one
"Translate all missing German translations" — triggers AI translation for every untranslated German key
"Delete all keys that start with 'old.'" — cleans up unused keys

The AI assistant figures out which tool to call and extracts the right parameters from your natural language. You never touch a form or click a button.

How It Works

The Verba MCP server is a lightweight TypeScript process that runs locally on your machine. Your AI assistant spawns it as a subprocess and communicates over stdin/stdout. The server translates your requests into Verba API calls.

You ↔ AI Assistant (Claude/Cursor) ↔ Verba MCP Server ↔ Verba API (verba.dev)

Your API key stays on your machine. The MCP server is open source so you can inspect exactly what it does.

Available Tools

Tool	What it does
`list_projects`	List all your Verba projects
`get_project`	Get project details — locales, key count, default language
`list_keys`	List and search translation keys with filtering
`add_key`	Create a new key with optional auto-translation
`set_translation`	Set a translation for a specific key and locale
`translate`	Trigger AI translation for one or more keys
`list_untranslated`	Show all keys missing translations
`add_locale`	Add a new language to your project
`delete_key`	Remove a key and all its translations

Setup

Claude Desktop

Add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "verba": {
      "command": "npx",
      "args": ["@verbadev/mcp-server"],
      "env": {
        "VERBA_API_KEY": "sk_your-secret-key"
      }
    }
  }
}

Restart Claude Desktop. You'll see the Verba tools available in the toolbar.

Claude Code

claude mcp add verba -- npx @verbadev/mcp-server

Set your API key, and you're ready to go.

Cursor / Windsurf

These editors support MCP servers through their settings. Add the Verba server with the same npx @verbadev/mcp-server command and your API key.

How Verba's Translation Workflow Has Evolved

Traditional TMS tools require you to leave your editor, open a dashboard, manually add keys, type default values, click translate, and switch back. For every single key. It's slow and breaks your flow.

Verba's SDK already fixed most of this. When you write verba.t('checkout.success', 'Payment successful!') in your code, the key gets created automatically with AI translations to all your configured languages. No dashboard required for day-to-day translation work.

Verba's MCP server closes the remaining gap. Tasks like reviewing untranslated keys, adding languages, bulk translating, or cleaning up old keys — things that still required the dashboard — now happen right in your editor through your AI assistant.

The progression: manual dashboard → code-first SDK → AI-assisted MCP. Each step removes more friction from the translation workflow.

Open Source

The Verba MCP server is open source under the MIT license. You can inspect the code, contribute improvements, or use it as a reference for building your own MCP server.

GitHub: github.com/verbadev/mcp-server

Getting Started

Sign up for Verba (free tier available)
Generate an API key in your dashboard settings
Install the MCP server using the instructions above
Start managing translations from your AI assistant

Verba (verba.dev) is a translation management system built for indie developers and small teams. Three lines of code to integrate, AI-powered translations, and OTA delivery — starting at $29/month for unlimited everything.