Tolgee

io.github.tolgee由 GitHub 帳號 tolgee 發布v1.0.0更新於 Oct 24, 2026

Your app's translations in Tolgee: search keys, create translations, trigger machine translation

已驗證Streamable HTTP可網頁執行效率與工作流開發者工具

概覽

AI 產生的概覽

Tolgee 託管的 MCP 伺服器讓 AI 助理管理在地化專案:搜尋與建立翻譯鍵、設定譯文並觸發機器翻譯。

可以做什麼
  • 列出與建立專案,查看各語言的翻譯進度統計
  • 列出、搜尋、建立、更新與刪除翻譯鍵,讀取鍵的中繼資料
  • 取得並設定鍵在各專案語言下的譯文
  • 列出與新增語言,列出命名空間,列出標籤並為鍵加上標籤
  • 為指定鍵啟動機器翻譯,並輪詢批次工作狀態
  • 列出、建立與刪除專案分支(需企業版分支功能)
適用情境
適合在使用 Tolgee 進行在地化的應用程式開發時,讓助理不必離開編輯器就能尋找翻譯鍵、補充或調整譯文、新增語言或標籤,或啟動機器翻譯。
執行需求
需要遠端 MCP 端點 app.tolgee.io(或自架執行個體)、可存取目標專案的 Tolgee 帳號,以及透過 X-API-Key 標頭傳入的 API 金鑰(個人存取權杖或專案 API 金鑰)。
安裝前請注意
此 API 金鑰可存取你的 Tolgee 專案,應採最小權限並妥善保管。刪除鍵與分支屬於破壞性操作,伺服器要求助理先與你確認。建立鍵、設定譯文、加標籤與機器翻譯都會變更專案資料,機器翻譯也可能消耗 Tolgee 中設定的付費翻譯服務額度。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Tolgee,將其新增到工作區。
  2. 連接 X-API-Key 憑證。它們會被私密儲存,絕不會在公開頁面上展示。
  3. 為需要使用其工具的對話啟用該服務。

Web executable,透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式,或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南,先向你展示發布者、端點和登入方式,經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.tolgee%2Ftolgee/install.md,並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前,先告訴我它的發布者、端點和登入方式,等我確認。不要把 API 金鑰寫進命令或我會提交的檔案;問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "tolgee": {
      "type": "http",
      "url": "https://app.tolgee.io/mcp/developer",
      "headers": {
        "X-API-Key": "<YOUR_API_KEY>"
      }
    }
  }
}

README

Tolgee
An open-source localization platform
developers enjoy working with

[Tolgee]

Why use Tolgee?

It saves you valuable time that would otherwise be spent on localization tasks while ensuring your software is flawlessly translated.

[Frame 47]

In context translating & One click screenshots

Easily add translations to your code and edit them directly within your app using the Tolgee i18n tool. Simply hold the ALT/Option key and click on an element to open a dialog where you can modify your strings effortlessly—no need to navigate through bulky .json/.po/.whatever files. Plus, in-context translation works seamlessly even in the production environment.

Just one click—that’s all it takes to capture a screenshot of your application with highlighted phrases for translation. Simply hold ALT + click on a string, then hit the camera button. Boom! Your screenshot is instantly generated.

[Tolgee in-context editing]

Translating on production

In-context translation works seamlessly even in the production environment of your deployed app. With the Tolgee Tools Chrome plugin, simply enter your API key and start translating—no coding skills required. This makes it easy for anyone to contribute to your app’s translations effortlessly.

True integrations

Tolgee isn't just another localization platform that simply syncs your local data with a backend. Instead, it seamlessly integrates into your app through powerful SDKs, providing a truly embedded translation experience.

Machine translation

We support DeepL, Google Translate, AWS Translate, and Azure Translator—simply choose your preferred services in the settings. With machine translation, the localization process becomes significantly faster, allowing translators to leverage AI-powered suggestions for more efficient and accurate translations.

AI translator

Conventional machine translation sees one string at a time. The AI translator sees the situation around it: screenshots of the phrase in your app, the key and project descriptions, your language notes, matches from the translation memory, and Tolgee Context, which the SDK gathers from how keys are actually laid out in your application.

The result is that ambiguous strings stop being a coin flip. A bare "Open" can be a button, a status or a menu item, and the AI translator picks based on where it appears—then tells you which reading it went with, so a reviewer can disagree at a glance.

You can shape it further by describing your project and languages, or by editing the translation prompt directly. Read more.

Translation memory

Tolgee automatically suggests translations based on those you've previously used in the project, ensuring consistency across similar phrases.

Translation memory suggestions include the similarity percentage, the key, and the original text of the translated string, making it easy to maintain accuracy and coherence.

Auto translation

When enabled, Tolgee instantly translates new keys using translation memory or machine translation services, ensuring your strings are translated immediately upon creation. Simply choose whether to use translation memory and select your preferred machine translation service for seamless automation.

Activity log

Easily track who modified, reviewed, or commented on phrases in your project—all with complete clarity.

Comment on translations

If something seems off, let others know what you'd change. You can leave comments on any translation directly within the Tolgee platform.

Translation history

Easily track changes to specific translations of a particular key in a specific language. If something's amiss, you'll know exactly where to point the finger!

MCP Server

Tolgee exposes an MCP (Model Context Protocol) server that lets AI coding assistants manage translations directly — search keys, create and update translations, manage languages, trigger machine translation, and more — all without leaving your editor. See DEVELOPMENT.md for setup instructions.

Documentation

The docs cover all aspects of using Tolgee. Some highlights include:

Getting Started 🚀

  1. Sign up at app.tolgee.io or access your self-hosted instance
  2. Create a project
  3. Follow a guide in the integration section of your project
  4. Have fun!

[Integration guides]

Report a Bug

For filing bugs, suggesting improvements, or requesting new features, please open an issue.

How to contribute

You wanna contribute? Great to hear that! You can start right away!

Contributors

[contributors]

Let us know what you think! #FeedbackWanted ❤️

來源:README.md,提交 1c6139c

工具

22

get_batch_job_status

Get the status of a batch job (e.g. machine translation) by its ID. Use to poll for completion.

輸入參數jobId必填projectId

machine_translate

Start machine translation for specified keys into target languages. Returns a batch job ID — use get_batch_job_status to poll for completion.

輸入參數branchkeyNames必填namespaceprojectIdtargetLanguageTags必填

store_big_meta

Store key relationships so Tolgee can use translations of related keys as context during machine translation, producing more consistent results. Keys that appear near each other in source code (e.g. on the same page or component) should be stored as related.

輸入參數branchprojectIdrelatedKeysInOrder必填

list_branches

List branches in a Tolgee project. Returns up to 100 results per page. Only available for projects with branching enabled (enterprise feature).

輸入參數pagesearchprojectId

create_branch

Create a new branch in a Tolgee project by forking from an existing branch. Only available for projects with branching enabled (enterprise feature).

輸入參數name必填projectIdoriginBranchId必填

delete_branch

Delete a branch from a Tolgee project. IMPORTANT: This is a destructive operation. The AI assistant should always confirm with the user before calling this tool.

輸入參數projectIdbranchName必填

list_keys

List translation keys in a Tolgee project. Returns up to 100 keys per page. Use search_keys to find specific keys by name or translation text.

輸入參數pagebranchprojectId

search_keys

Search for translation keys in a Tolgee project by name or translation text. Returns up to 50 matching keys per page. Note: namespace and tags filtering is not yet supported — filter results client-side if needed.

輸入參數pagequery必填branchprojectIdlanguageTag

create_keys

Create translation keys in a Tolgee project with optional translations and tags. Keys that already exist are silently skipped — their translations and tags are not updated. Use update_key and set_translation to modify existing keys.

輸入參數keys必填branchnamespaceprojectId

get_key

Get a translation key's metadata (name, namespace, description) by its name. To get the key's translations, use get_translations.

輸入參數branchkeyName必填namespaceprojectId

update_key

Update an existing translation key's name, namespace, or description

輸入參數keyName必填newName必填keyBranchprojectIdkeyNamespacenewNamespacenewDescription

delete_keys

Delete translation keys from a Tolgee project. IMPORTANT: This is a destructive operation. The AI assistant should always confirm with the user before calling this tool.

輸入參數branchkeyNames必填namespaceprojectId

list_languages

List all languages configured for a Tolgee project. Returns up to 1000 results per page.

輸入參數pageprojectId

create_language

Add a new language to a Tolgee project

輸入參數tag必填name必填flagEmojiprojectIdoriginalName

list_namespaces

List all namespaces in a Tolgee project. Namespaces organize translation keys into logical groups (e.g. by feature, page, or module).

輸入參數projectId

list_projects

List Tolgee projects accessible to the current user. Returns up to 100 results per page.

輸入參數pagesearch

create_project

Create a new Tolgee project with initial languages

輸入參數name必填languages必填organizationId必填baseLanguageTag

get_project_language_statistics

Get translation status and progress for each language in a project. Returns per-language statistics including translated, reviewed, and untranslated percentages.

輸入參數projectId

list_tags

List all tags used in a Tolgee project. Returns up to 1000 results per page.

輸入參數pagesearchprojectId

tag_keys

Add tags to translation keys. Creates tags if they don't exist.

輸入參數tags必填branchkeyNames必填namespaceprojectId

get_translations

Get translations for a specific key in a Tolgee project. Returns translations in all project languages, or only the specified languages if provided.

輸入參數branchkeyName必填languagesnamespaceprojectId

set_translation

Set or update translations for a key in one or more languages. The key must already exist — use create_keys to create it first.

輸入參數branchkeyName必填namespaceprojectIdtranslations必填

版本歷史

1
  1. v1.0.0最新Sep 16, 2026

評分

已滿足適用檢查項的 90%

該評分衡量此收錄條目的完整程度,依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價,也不表示伺服器安全或目前可用:可連通性是帶日期的檢查結果,評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分

    請在儲存庫或註冊中心條目中宣告授權條款。