globalize.now

Official remote MCP server for globalize.now, AI-native localization infrastructure.

От сообществаРаботаетБез входаГлобальныйБесплатноТолько чтение

Что умеет

  • List Languages: List the global Globalize language catalog. Optionally filter with a search term (matches language name or locale). Returns each language with the id and locale needed to create a proj
  • List Projects: List all projects in your organisation, each with its source and target languages. When referring to a project or language in your reply to the user, use its name and locale (e.g. "Fren
  • Get Project: Get a single project by its id or slug, including its languages and connected repository. When referring to a project or language in your reply to the user, use its name and locale (e.g.

Какие данные видит

Нужен ли аккаунт

Не нужен: сервер работает без входа

Official remote MCP server for globalize.now, AI-native localization infrastructure. Translate locale files (JSON, PO, XLIFF, YAML, Android XML, iOS xcstrings, Markdown, HTML), manage glossaries, per-language style guides and project context, and connect GitHub repositories so translations stay in your repo as locale files. 21 tools. Works in Claude (listed in the connector directory), Cursor and any MCP client with OAuth.

Как подключить

Как подключить globalize.now к Claude

Адрес сервера, который нужно вставить в настройках Claude:

https://globalize--globalize-now.run.tools

  1. Откройте Claude (сайт claude.ai или приложение для компьютера).
  2. Перейдите в Настройки → Коннекторы (в английском интерфейсе: Settings → Connectors).
  3. Нажмите Добавить пользовательский коннектор (Add custom connector).
  4. Вставьте адрес сервера, скопированный ниже, в поле URL удалённого MCP-сервера и нажмите Добавить.
  5. В чате нажмите + → Коннекторы и включите новый коннектор.

Пользовательские коннекторы доступны на тарифах Free, Pro, Max, Team и Enterprise. На Free можно добавить только один. В Team и Enterprise коннектор сначала добавляет владелец организации в Настройки организации → Коннекторы.

Авторизация

После нажатия Добавить откроется окно входа в сервис. Войдите под своим аккаунтом и подтвердите доступ. Пароль Claude не видит.

Как подключить globalize.now к ChatGPT

Адрес сервера, который нужно вставить в настройках ChatGPT:

https://globalize--globalize-now.run.tools

  1. Откройте ChatGPT в браузере (chatgpt.com). Нужен тариф Plus, Pro, Business, Enterprise или Edu.
  2. Один раз включите режим разработчика: Настройки → Приложения → Дополнительные настройки → Режим разработчика (Settings → Apps → Advanced settings → Developer mode).
  3. Откройте Настройки → Коннекторы и нажмите Создать (Create).
  4. Заполните форму: Название (любое), Описание (одна фраза, что делает сервис), URL MCP-сервера (скопируйте ниже).
  5. В поле Аутентификация выберите OAuth, если сервис требует вход, иначе Нет. Нажмите Создать.
  6. В новом чате откройте + → Приложения/Коннекторы и включите коннектор.

Название раздела у OpenAI менялось (Connectors → Apps/Plugins). Если пункт называется иначе, ищите «режим разработчика» в настройках. В рабочих пространствах Business/Enterprise администратор должен разрешить пользовательские коннекторы.

Авторизация

При первом использовании ChatGPT откроет окно входа в сервис. Войдите и подтвердите доступ.

Как подключить globalize.now к Cursor

Адрес сервера, который нужно вставить в настройках Cursor:

https://globalize--globalize-now.run.tools

Быстрый способ: нажмите кнопку Открыть в Cursor ниже, Cursor сам предложит добавить сервер.

Вручную:

  1. В Cursor откройте Settings → Cursor Settings → MCP и нажмите Add new global MCP server.
  2. Вставьте JSON, скопированный ниже, в открывшийся файл ~/.cursor/mcp.json (для одного проекта: .cursor/mcp.json в корне проекта).
  3. Сохраните файл. Сервер появится в списке MCP; при необходимости пройдите авторизацию по кнопке рядом с ним.

Авторизация

Если сервис требует вход, рядом с сервером в списке MCP появится кнопка авторизации.

Список инструментов сервера (21)

Технические названия из tools/list. Нужны только разработчикам.

list_languagesList the global Globalize language catalog. Optionally filter with a search term (matches language name or locale). Returns each language with the id and locale needed to create a project or add a project language. When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.
list_projectsList all projects in your organisation, each with its source and target languages. When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.
get_projectGet a single project by its id or slug, including its languages and connected repository. When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.
create_projectCreate a localisation project from catalog languages. Resolve language ids first with list_languages. The source language is what your content is authored in; target languages are what it will be translated into. When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.
add_project_languageAdd a catalog language as a new target language on an existing project. Resolve the catalog language id first with list_languages. When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.
remove_project_languageRemove a target language from a project. Pass the project_language id (from the targetLanguages list returned by get_project), not a catalog language id. When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.
list_github_installationsList the GitHub App installations available to your organisation. Use the returned numeric installationId with list_repository_branches, detect_repository, and connect_repository.
start_github_installBegin connecting GitHub. Returns an install URL — surface it to the user to open in a browser and install the Globalize GitHub App. After they confirm, call list_github_installations to get the installation id.
list_repository_branchesList the branches of a GitHub repository accessible to an installation. Use the numeric installationId from list_github_installations.
detect_repositoryScan a GitHub repository to detect its framework, locale path patterns, source/target languages, and file format. Use the result to inform create_project and set_repository_patterns. Use the numeric installationId from list_github_installations.
connect_repositoryConnect a Git repository to a project, optionally with locale path patterns. For GitHub, pass the githubInstallationId from list_github_installations. Registers webhooks and runs best-effort namespace discovery.
set_repository_patternsAdd locale path patterns to a connected repository (existing identical patterns are skipped). Each pattern must contain exactly one {locale} and at most one {namespace}.
translate_filesSubmit one or more source files to be translated as a single job. Each file is parsed by its content (JSON — flat, nested, ARB and Chrome/WebExtension messages.json — plus PO, XLIFF 1.2 and 2.0, YAML, @wxt-dev/i18n TOML/JSON5/JSONC catalogs, Android XML, iOS xcstrings, Markdown and HTML), so pass the file as-is. Target languages default to the project's configured target languages; pass a subset to translate only some. Returns { jobId, status }. Translation runs asynchronously. This call returns immediately with a jobId; it does NOT wait for translation to finish. Call get_translation with the jobId and keep calling it until status is "completed" (or a terminal failure), then read the files.
get_translationPoll a translation job started by translate_files and, once it is "completed", pull back the translated files. While the job is still running, returns its status and progress with no file content — call again until status is "completed". A "completed" response carries each translated file (one per namespace × target language) plus a manifest; a "failed" response carries the failure reason. Optionally filter by target language to bound the payload of a large job.
list_glossaryList the glossary entries for a project — each a source→target term pair with its languages and id. Optionally filter by source and/or target project_language id. Results are paginated; pass the returned cursor to fetch the next page. When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.
add_glossary_entriesAdd (or update) glossary entries for a project in one call. Pass an array of term rows, each with its own source and target term and their project_language ids — so a single source term can have a different translation per target language. Resolve the language ids first with get_project. Entries are upserted on (project, sourceTerm, source, target); re-adding an existing key updates its target term and its do-not-translate flag. Set doNotTranslate on a row for a term that must be kept verbatim (a brand or product name); its stored target term is then forced to equal the source term. Marks any matching translation-memory entries stale so future translations pick up the change. When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.
delete_glossary_entryDelete a single glossary entry by its id (from list_glossary) from a project in your organisation. Marks matching translation-memory entries stale so the affected content is re-translated without the removed term. When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.
list_style_guidesList the per-language style guides for a project — each the free-text translation instructions for one target language, with that language and the guide id. When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.
set_style_guideCreate or replace the style guide (free-text translation instructions) for one target language of a project, keyed by its project_language id (from get_project). You draft the instructions yourself. A brand-new guide always marks that language's translation memory stale so future translations follow it; for an update, set recheckExistingTranslations to also re-check already-translated content (it defaults to off). When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.
delete_style_guideDelete the style guide for one target language of a project, keyed by its project_language id (from get_project or list_style_guides). When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.
set_project_contextSet a project's project-wide context — the free-text prompt guidance shared across every target language (what the product does, its audience, tone). At most 1000 characters. Read the current value with get_project. When referring to a project or language in your reply to the user, use its name and locale (e.g. "French (fr-FR)") — ids (UUIDs) are for tool calls only, never show them to the user.