You can run a Smartling MCP Server to expose Smartling translation tools to Claude Desktop, Cursor, and other MCP-compatible apps. This server provides secure access to translation projects, files, glossaries, contexts, and more, enabling you to manage translations directly from your MCP clients.

How to use

You connect your MCP client to this server by starting the local Smartling MCP service and pointing your client to the local process. Once running, your MCP client can access a suite of translation tools including project management, file operations, job management, string operations, and quality assurance. Use the server to upload translation assets, manage locales, push context for translators, trigger translation jobs, and retrieve progress reports. You can also integrate with internal chat platforms for real-time notifications and workflows.

Typical usage patterns include starting the server, authenticating with your Smartling API credentials, and then issuing MCP commands from Claude Desktop or Cursor to create translation jobs, upload files, manage glossaries, and fetch translation statuses. If you need to verify the connection, ask your MCP client to report the available Smartling tools and their counts. This confirms that the tools are loaded and ready for use.

How to install

Prerequisites: You need Node.js 18.0.0 or higher and Smartling API credentials (User Identifier and User Secret). Ensure your environment has network access to Smartling API endpoints.

Step 1: Clone the server repository and install dependencies.

# Clone the repository
git clone https://github.com/YOUR_USERNAME/smartling-mcp-server.git
cd smartling-mcp-server

# Install dependencies
npm install

# Build the project
npm run build

Step 2: Set up environment variables for credentials and API access.

# Copy example config and edit credentials
cp config-example.env .env

Then edit the .env file to include your Smartling credentials and API base URL. The required variables are shown in the example configuration.

Step 3: Add your credentials to the configuration files.

Claude Desktop example path

Edit: ~/Library/Application Support/Claude/claude_desktop_config.json

Cursor example path

Edit: ~/.cursor/mcp.json

Replace the credential placeholders with your actual values in the appropriate configuration files for Claude Desktop and Cursor as described in the setup notes.

Step 4: Obtain Smartling credentials.

• Go to Smartling Dashboard → Settings → API
• Create User Identifier and User Secret
• Copy them into your config files
• Restart Claude Desktop and Cursor after updating credentials

Configuration example

Below is the runtime configuration for the Smartling MCP Server when used with an MCP client. The server runs locally via Node.js and uses a path to the built distribution.

{
  "mcpServers": {
    "smartling": {
      "command": "node",
      "args": ["/path/to/smartling-mcp-server/dist/index.js"],
      "env": {
        "SMARTLING_USER_IDENTIFIER": "your_user_identifier",
        "SMARTLING_USER_SECRET": "your_user_secret",
        "SMARTLING_BASE_URL": "https://api.smartling.com"
      }
    }
  }
}

Troubleshooting

If you encounter issues, verify that the server process is running and that your credentials are correct. Common problems include authentication failures, invalid API keys, and network timeouts. Check that the Node.js runtime is version 18 or newer and that the environment variables are loaded by the process. Restart the clients after credential updates.

Notes on security and maintenance

Keep your Smartling credentials secure. Do not expose the User Identifier or User Secret in client-side configurations or logs. Rotate credentials periodically and use scoped API access where available. Regularly rebuild and restart the server after credential changes or Smartling API updates.

Available tools

Project Management

Access and manage translation projects within Smartling via MCP, including creating projects and organizing workflows.

File Operations

Upload, download, and manage translation files and assets used in localization projects.

Job Management

Create and monitor translation jobs, assign tasks, and track progress.

String Operations

Search, filter, and manage translation strings across locales.

Quality Assurance

Run translation quality checks and validate linguistic accuracy.

Locales

Manage project locales and supported languages for translations.

Glossaries

Manage glossaries and terminology for consistent translations.

Context

Upload and manage visual context to assist translators.

Webhooks

Set up notifications for translation events and project changes.

Reports

Generate progress and analytics reports on translation activities.

Internal Platform Client

Node.js backend integration for internal chat platforms.

Streaming Client

WebSocket integration for live streaming interactions with MCP.

Usage Examples

Provide integration examples and usage patterns for developers.