Intento Translate MCP Server

What is Intento Translate MCP Server about?

The server implements an MCP endpoint that allows clients to translate text between any languages supported by the Intento API. It also publishes a resource containing the latest mapping of language names to ISO codes, keeping language support up‑to‑date automatically.

How to use Intento Translate MCP Server?

1. Install dependencies

   pip install -r requirements.txt
   

2. Configure the Intento API key
   • Create a .env file in the project root with INTENTO_API_KEY=your_api_key or specify the key in the env section of your MCP client configuration.
3. Start the server

   python server.py
   

4. Call the translate tool

   result = await translate("Hello world", target_language="Spanish")
   # or with explicit source language
   result = await translate("Hello world", target_language="es", source_language="en")
   

5. Fetch the language‑codes resource

   language_codes = await get_resource("mcp://intento-translate/language-codes")
   spanish_code = language_codes["codes"]["spanish"]
   

Key features of Intento Translate MCP Server

• Dynamic language list – pulls the latest language‑code mapping from Intento at runtime.
• Automatic source‑language detection – omits the source field when not provided, letting Intento detect the language.
• Both names and ISO codes accepted – flexibility for agents that prefer human‑readable names or strict codes.
• Robust error handling & logging – all activity recorded in logs/intento-translate.log with clear messages for API failures and invalid codes.
• Easy integration with Claude Desktop – ready‑to‑use MCP config example included.

Use cases of Intento Translate MCP Server

• Multilingual chatbot assistants that need on‑the‑fly translation of user inputs and responses.
• Content localization pipelines where agents fetch the latest language codes and translate strings programmatically.
• Cross‑language data enrichment for analysis tools that ingest text in various languages.
• Integration with AI agents (e.g., Claude Desktop) to provide seamless translation services without hard‑coding language mappings.

FAQ from the Intento Translate MCP Server

Q: Do I need to specify a source language? A: No. Leaving source_language empty triggers automatic detection by the Intento API.

Q: How often are language codes refreshed? A: The server fetches them each time it starts; you can restart the server to pick up any new languages added by Intento.

Q: What happens if the Intento API call fails? A: The server returns a descriptive error message and falls back to a minimal common‑language set for continued operation.

Q: Can different MCP clients use different API keys? A: Yes. Set the INTENTO_API_KEY in the env section of each client’s MCP config JSON to override the server‑wide key.

Q: Where are logs stored? A: All logs are written to logs/intento-translate.log under the project directory.

Intento Translation MCP Server

This is an MCP server that provides translation capabilities using the Intento API.

Setup

1. Install the required dependencies:

pip install -r requirements.txt

2. Set up your Intento API credentials

There are two options:

2.a. Setting Intento API credentials for the MCP server (all your MCP clients will use the same credentials)

To set up Intento API credentials for the MCP server:

• Create a .env file in the project root
• Add your Intento API key:

INTENTO_API_KEY=your_api_key_here

2.b. Setting Intento API credentials for the specific MCP integration, say, Claude Desktop (different MCP clients may use different credentials)

To set up Intento API credentials for the specific MCP integration:

• Use the "env" section of your MCP config JSON file to specify the "INTENTO_API_KEY" variable. See the Claude Desktop Integration part as an example

Usage

1. Start the server:

python server.py

2. The server provides the following tools and resources:

translate

Translates text between languages. Accepts both language names and ISO codes.

Parameters:

• text (required): The text to translate
• target_language (required): Target language (e.g., "Spanish" or "es")
• source_language (optional): Source language (defaults to empty string for automatic detection)

Example:

# Translate with automatic language detection
result = await translate("Hello world", target_language="Spanish")

# Translate with specific source language
result = await translate("Hello world", target_language="Spanish", source_language="English")

language-codes (Resource)

Resource URL: mcp://intento-translate/language-codes

Provides a mapping of language names to their ISO codes as supported by the Intento API.

This resource allows agents to:

• Discover available languages
• Look up the correct ISO code for a language name
• Ensure compatibility with the Intento API

Example of using the resource:

# Get the language codes resource
language_codes = await get_resource("mcp://intento-translate/language-codes")

# Look up a language code
spanish_code = language_codes["codes"]["spanish"]  # Returns "es"

# Use the code in a translation
result = await translate("Hello world", target_language=spanish_code)

Language Support

The server dynamically fetches language codes from the Intento API and makes them available as a resource. This approach has several advantages:

1. Dynamic Updates: Language support is always up-to-date with the Intento API
2. Agent Autonomy: Agents can discover and use the correct language codes themselves
3. Flexibility: The server can handle both language names and ISO codes

Automatic Language Detection

When the source_language parameter is empty (the default), the server will:

• Omit the source language field in the API request
• Let Intento automatically detect the source language
• Log the request as using "auto-detect" for clarity

If the API request fails, the server falls back to a minimal set of common language codes.

Error Handling

The server includes comprehensive error handling and logging:

• All operations are logged to logs/intento-translate.log
• API errors are caught and returned with descriptive messages
• Invalid language codes are handled gracefully

Claude Desktop Integration

To use this MCP server with Claude Desktop:

1. Create a claude-mcp-config.json file in your Claude Desktop configuration directory:

   • Windows: %APPDATA%\Claude\claude-mcp-config.json
   • macOS: ~/Library/Application Support/Claude/claude-mcp-config.json
   • Linux: ~/.config/Claude/claude-mcp-config.json

2. Add the following configuration (adjust paths to match your installation):

{
    "mcpServers": {
        "intento-translate": {
            "command": "python",
            "args": ["/path/to/your/mcp-intento-translate/server.py"],
            "cwd": "/path/to/your/mcp-intento-translate",
            "env": {
                "INTENTO_API_KEY": "<YOUR_API_KEY>" 
            }
        }
    }
}

The "env" part with "INTENTO_API_KEY" is required only if you want to use specific credentials for the integration and not use MCP server-specific credentials.

3. Restart Claude Desktop to load the new configuration.

The server will now be available to Claude Desktop, providing translation capabilities and language code information through the MCP protocol.