deepl

This MCP server exposes five tools that wrap the DeepL API for language translation and related capabilities: translate_text, rephrase_text, get_source_languages, get_target_languages, and get_usage. You can call these tools via your MCP client by sending the appropriate JSON payloads to the server. Core features include formality control, context hints, glossary support, batch translation, and automatic detection of free vs. pro API keys to route requests to the correct DeepL endpoint. To get started, run the server with Node.js and point your MCP client to the built index.js, ensuring your DeepL API key is provided via environment variables.

Useful capabilities include:

• translate_text: translate text with target_lang, optional formality, context, and glossary support. You can also translate arrays of texts in a single request (batch translation).
• rephrase_text: use DeepL Write (Pro only) to improve clarity and style for a given text.
• get_source_languages / get_target_languages: retrieve available languages and formality support information.
• get_usage: obtain current character usage, limits, and API type (Free vs Pro).

Configuring your MCP client to use these tools typically involves wiring your JSON requests to the server’s index.js entry point and including DEEPL_API_KEY in the environment so the server can authenticate with DeepL.

Prerequisites:

• Node.js 18+ installed
• npm (comes with Node.js) or pnpm/yarn if preferred

Installation steps:

1. Clone the repository: git clone https://github.com/watchdealer-pavel/deepl-mcp-server.git
2. Navigate to the project directory: cd deepl-mcp-server
3. Install dependencies: npm install
4. Build the server (if a build step is required by the project): npm run build
5. Run the server: npm start

Environment configuration:

• Ensure you have a DeepL API key. The server expects DEEPL_API_KEY to be set in the environment when running.
• If you’re using a local MCP client like Claude, configure the client to point at the server’s built index.js as shown in the Quick Start section of the README.

Tips and common considerations:

• The server auto-detects Free vs Pro API keys by key format (ends with :fx for Free; others are Pro). Ensure you provide the correct key for your usage scenario.
• When translating multiple texts, you can leverage batch translation to reduce latency and API calls.
• For rephrase_text (DeepL Write), note that this feature requires a Pro API key.
• If you encounter authentication errors, verify that DEEPL_API_KEY is correctly exported in the environment where the Node process runs.
• The MCP client configuration examples include the required command and arguments. Adapt the /path/to/deepl-mcp-server/build/index.js path to your actual deployment path.
• You can specify additional metadata or default options in your MCP client configuration if needed (for example, default target language or formality).