What is the Model Context Protocol (MCP)?

The Model Context Protocol (MCP) is an open standard that enables AI models to securely connect to external tools and data sources. It acts as a universal bridge, standardizing how AI assistants communicate with APIs like Lara Translate.

Which AI assistants and IDEs are supported?

Because Lara MCP is built on the open standard, it works out of the box with any MCP-compliant client. This currently includes Anthropic's Claude Desktop app, modern code editors like Cursor and Zed, and an ever-growing list of tools.

Is the MCP server free to use?

The MCP server itself is free and open-source. However, when the AI assistant uses the tool to translate text, it will authenticate with your Lara API Key and consume characters from your standard API quota or pay-as-you-go plan.

MCP Server for AI Translation

The first purpose-built MCP server for language translation. Bring enterprise-grade, context-aware translation into Claude Desktop, Cursor, Windsurf, Cline, and any MCP-compatible IDE — across 200+ languages, with no custom API integration required.

mcp_client_config.json

Read MCP Documentation

The Universal Bridge for AI Translations

The Model Context Protocol (MCP) lets you turn Lara Translate into a native tool for your favorite AI assistants. Let them translate code, documents, and chats automatically with full context.

Universal AI Compatibility

Native integration with any MCP-compliant application. Just add the Lara server to your tool's config file and ask your AI to translate instantly.

Context Protocol Standard

Built on the open Model Context Protocol. Future-proof your integration and use Lara with emerging AI clients without rewriting custom API calls.

AI-Driven Disambiguation

Because the AI has access to your local files or chat context, it automatically feeds relevant contextual metadata to Lara, ensuring perfectly accurate translations.

Zero-Data Training Policy

Your translations are routed securely to the Lara API. We enforce strict data privacy, meaning your proprietary strings are never used to train public models.

Access 200+ Languages

Give your AI the power to understand and generate localized content in over 200 languages with production-grade reliability.

Format Handling

Because it's integrated via MCP, your AI can read your HTML, Markdown, or JSON, ask Lara to translate the raw strings, and cleanly rewrite the file.

How to connect Lara Translate to your AI assistant or IDE

One config block. Works with Claude Desktop, Cursor, Windsurf, Cline, and any MCP-compatible client.

Get your Lara API credentials

Sign up for a free account or find them in your Lara profile. You'll need your LARA_ACCESS_KEY_ID and LARA_ACCESS_KEY_SECRET in the next step.

Configure your MCP Client

Open the JSON configuration file for your chosen tool (e.g. Claude Desktop) and add the Lara server mapping.

"lara-translate": {
"command": "npx",
"args": [
"-y",
"@translated/lara-mcp@latest"
],
"env": {
"LARA_ACCESS_KEY_ID": "<LARA_ACCESS_KEY_ID>",
"LARA_ACCESS_KEY_SECRET": "<LARA_ACCESS_KEY_SECRET>"
}
}

Start Translating

Restart your client application. You will now see the Lara Translate tool available in your AI's context window. Just prompt: “Translate this file to French”.

Explore all available tools: translate, list_memories, create_memory, list_glossaries in the Lara MCP documentation.