概览
What is Lara Translate MCP Server?
A Model Context Protocol (MCP) server for Lara Translate, enabling professional translation capabilities with support for language detection, context-aware translations, translation memories, and glossaries. It leverages Translation Language Models (T-LMs) trained on billions of professionally translated segments, delivering domain-specific translations that capture cultural nuances and industry terminology that general-purpose LLMs often miss.
How to use Lara Translate MCP Server?
Add the server URL (https://mcp-v2.laratranslate.com/v1) to your MCP client config — the client handles OAuth authentication automatically. For clients like Claude Desktop, Cursor, and Claude Code, you can add it as a custom connector or install from the plugin marketplace. Alternatively, self-host using NPX, Docker, or from source with access key credentials. Test with a prompt like Translate with Lara "Hello world" to Spanish.
Key features of Lara Translate MCP Server
- Professional translation with context, instructions, memories, and glossaries
- Multiple translation styles: faithful, fluid, and creative
- Language detection and listing of supported languages
- Full CRUD management of translation memories and TMX imports
- Full CRUD management of glossaries with CSV import/export
- OAuth 2.0 and access key authentication options
Use cases of Lara Translate MCP Server
- Translate text with domain-specific accuracy using translation memories and glossaries
- Detect the language of input text or arrays of text
- Manage translation memories: create, update, delete, add translations, import TMX files
- Manage glossaries: create, update, delete, add terms, import/export CSV files
- Integrate professional translation capabilities into AI assistants and development tools
FAQ from Lara Translate MCP Server
What authentication methods are supported?
OAuth 2.0 (default) — no API keys needed, your client handles browser login automatically. Alternative: access key authentication with an Access Key ID and Secret.
What are the runtime requirements for self-hosting?
Self-hosting via NPX requires Node.js. Self-hosting via Docker requires Docker. Building from source requires Node.js and pnpm.
Where does the translation data live?
Translation memories and glossaries are managed within your Lara Translate account. All data lives on Lara Translate's servers; the MCP server acts as a bridge to those services.
What are the known limitations?
The README does not specify any known limitations.
What transport protocols are supported?
The README demonstrates the hosted endpoint using HTTP (server URL), and self-hosting using STDIO (via NPX, Docker, or built from source).
