Verblaze

Provides programmatic access to Verblaze localization data via MCP for language management, screens, and translations.

javascript

0

GitHub Stars

javascript

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "verblaze-verblaze_mcp_server": {
      "command": "npx",
      "args": [
        "-y",
        "verblaze-mcp-server@latest"
      ],
      "env": {
        "VERBLAZE_API_KEY": "<your-api-key>"
      }
    }
  }
}

You connect Verblaze localization projects to AI tools using MCP to enable assistants to manage languages, screens, and translations on your behalf. This MCP server acts as a bridge between Verblaze and your preferred AI tools, letting you drive project work with prompts from your trusted assistants while keeping authentication and project scope under your control.

How to use

Use the Verblaze MCP Server to enable your AI tools to interact with your localization projects. Once connected, your assistant can translate values, add or remove languages, manage screens, and retrieve translations across languages. You can operate in a project-scoped mode so the assistant only touches the specific project you authorize.

To start, ensure you have your Verblaze API key and a compatible MCP client installed. Connect the MCP server from your chosen tool by providing the Verblaze MCP server configuration and your API key. After a successful connection, verify the server status in the client’s MCP settings and begin issuing prompts like “list all languages,” “add a language,” or “get translations for screen X.” Remember to review tool calls before execution to protect your data.

How to install

Prerequisites: Node.js (for runtime and npm/npx) and a Verblaze project with an API key.

Prepare your environment by installing Node.js if it is not already installed.
Install or run the MCP server using your chosen MCP client’s recommended flow. The Verblaze MCP server is exposed to clients as a Node-based executable that you run with npx.
Provide your Verblaze project API key as an environment variable to authorize requests.
Start using your MCP-enabled tool to connect to Verblaze and begin issuing localization tasks through natural prompts.

Security and best practices

Understand the risks of connecting an LLM to your Verblaze data. Follow security best practices to minimize exposure, review tool calls before execution, and scope access to a specific project.

Key recommendations include avoiding production data during experiments, keeping tool access internal to developers, using project-scoped configurations, and preferring read-only operations when exploring data before making changes.

Prompt injection is a primary concern. Always enable manual review of tool calls and confirm the details of each action before proceeding.

Notes on configuring the MCP server with your tools

The following example shows how to configure a single Verblaze MCP server instance for use with your MCP-enabled tools. This server uses npx to launch the Verblaze MCP server and requires your Verblaze API key.

Next steps

Your AI tool is now connected to Verblaze using MCP. Try prompts like: list all languages, add a new language, get translations for a screen, translate values to a new language, or manage project users and permissions.

Security risks

Connecting any data source to an LLM carries inherent risks. You should evaluate risks such as prompt injection and data exposure and implement safeguards accordingly.

Recommendations include using development projects, restricting access to developers only, scoping to a single project, reviewing tool calls before executing destructive actions, and preferring read-only exploration when possible.

Available tools

translateValues

Translate string values to all supported languages in a project. Requires fileKey and values object.

addLanguage

Add a new language to the project. Requires languageCode (e.g., 'en-US', 'es-ES').

removeLanguage

Remove a language from the project. Requires languageCode.

changeBaseLanguage

Change the base language of the project. Requires languageCode.

listLanguages

List all supported languages in the project. No parameters required.

addScreen

Add a new screen/file to all languages. Requires fileKey or fileTitle.

removeScreen

Remove a screen/file from all languages. Requires fileKey.

listScreens

List all screens/files in the project. No parameters required.

deleteValue

Delete a specific translation value from all languages. Requires fileKey and valueKey.

getScreenValues

Get all translation values for a specific screen in a specific language. Requires languageCode and fileKey.