mcpField
T
translatize

translatize

Local
github.com/TranslatizeDeveloper Tools0Updated 1w ago
Repository

About translatize

translatize is a set of TypeScript tools for working with the Translatize integration API. It includes an SDK, a CLI for translation sync and branching, and an MCP server for AI agents.

Key features

  1. 01TypeScript SDK for the integration API
  2. 02Read, search, edit, and branch labels
  3. 03CLI for translation pull, push, and status
  4. 04MCP server for AI agents
  5. 05CI templates for GitHub and GitLab

Use cases

  1. 01Read and update translation labels
  2. 02Sync translations from a CI workflow
  3. 03Expose translation API operations to AI agents

MCP support

The @translatize/mcp package provides an MCP server that exposes the integration API to AI agents over stdio.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "com-translatize": {
      "command": "npx",
      "args": [
        "-y",
        "@translatize/mcp"
      ],
      "env": {
        "TRANSLATIZE_API_TOKEN": "<TRANSLATIZE_API_TOKEN>",
        "TRANSLATIZE_API_URL": "<TRANSLATIZE_API_URL>",
        "TRANSLATIZE_APP_URL": "<TRANSLATIZE_APP_URL>"
      }
    }
  }
}
Run commandnpx -y @translatize/mcp
Environment variables to set
TRANSLATIZE_API_TOKENSecretTranslatize integration API token (an "mcni_..." string), bound to one project and branch. Create it under Project Settings -> Integrations at https://app.translatize.com.
TRANSLATIZE_API_URLAPI base URL including the /v1 segment. Defaults to https://api.translatize.com/v1.
TRANSLATIZE_APP_URLWeb-app base URL used to build review links. Defaults to https://app.translatize.com.

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@translatize/mcp0.1.3

Tools

18

Saved tool list · updated 1w ago

This tool takes no parameters.

Details

From the project README

Translatize integrations

Client-side tooling for the Translatize integration API (the machine-token API authenticated with Authorization: Bearer mcni_...). This is an npm workspace root that is intentionally separate from the apps in the repo – it has the only package.json with workspaces, and the repo root stays package.json-free.

Packages

PackageDirectoryStatusWhat it is
@translatize/corecore/availableZero-dependency TypeScript SDK for the integration API (client + flatten/unflatten helpers). The shared foundation everything else builds on.
@translatize/clicli/availableCommand-line tool (translatize init/pull/push/status) built on @translatize/core. Syncs translations from CI with git-like branching.
@translatize/mcpmcp/availableModel Context Protocol server exposing the API to AI agents (Claude Code, Cursor, ...) over stdio, built on @translatize/core.
CI templates

Two ready-to-use CI integrations complement the packages. They shell out to the @translatize/cli (npx @translatize/cli ...) rather than being npm packages, so you consume them as workflow/pipeline config, not via npm install:

TemplateHomeWhat it is
GitHub Actiongithub.com/Translatize/sync-actionComposite action (translatize/sync-action) wrapping the CLI's pull/push/status for GitHub workflows.
GitLab CIgitlab.com/translatize/gitlabCI/CD Catalog component (templates/translatize.yml) plus copy-paste .gitlab-ci.yml jobs.

Development

cd integrations
npm install        # installs every workspace and hoists dev deps
npm run build      # builds each workspace that defines a build script
npm test           # runs each workspace's test suite

Requires Node.js >= 18.17 (native fetch, node:test).

Publishing

The @translatize/core, @translatize/cli, and @translatize/mcp packages are published to npm under the @translatize scope. Publishing is manual – nothing here publishes automatically. To cut a release, build first, then npm publish --access public from the individual package directory (or npm publish --workspaces).

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.