Google Translate MCP server for AI agents
Let Claude, ChatGPT or your own agent read and act on Google Translate: 18 typed functions through one MCP server. No Google Translate API key needed.
- Functions
- 18
- Sign-in
- For account actions
- Works with
- MCP, HTTP, Node, Python
Connect
An MCP server for Google Translate alone, with one tool per function. Paste it into Claude, ChatGPT, Cursor or any MCP client as a remote server.
https://api.bowmark.ai/mcp/site/google_translateOne MCP server for Google Translate and every other site Bowmark covers. Your agent looks up what it needs and calls it.
https://api.bowmark.ai/mcpSet it up in your clientBoth need a Bowmark API key. The quickstart gets you one and a first result in a few minutes. Functions that need a login use the person's own account: how sign-in works.
Functions
Open a function to see its arguments, what it returns and how to call it.
checkSpellingGoogle Translate's own "Did you mean …" line.Arguments
Name Type Required textstringYes languagestringYes Returns
GoogleTranslateSpellCheckinterface GoogleTranslateSpellCheck { text: string; language: string; correct: boolean; corrected?: string; correctedHtml?: string; correctionType?: number[]; confident?: boolean; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/checkSpellingCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.checkSpelling({ text: "I wnat to eat diner", language: "English" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.checkSpelling({"text": "I wnat to eat diner", "language": "English"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/checkSpelling \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"text":"I wnat to eat diner","language":"English"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.checkSpelling({ text: "I wnat to eat diner", language: "English" });detectLanguageWork out what language a string is written in, with Google's own confidence in the answer and the other languages it considered.Arguments
Name Type Required textstringYes Returns
GoogleTranslateLanguageDetectioninterface GoogleTranslateLanguageCandidate { language: string; confidence: number; } interface GoogleTranslateLanguageDetection { language: string; confidence: number; candidates: GoogleTranslateLanguageCandidate[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/detectLanguageCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.detectLanguage({ text: "Bonjour tout le monde" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.detectLanguage({"text": "Bonjour tout le monde"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/detectLanguage \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"text":"Bonjour tout le monde"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.detectLanguage({ text: "Bonjour tout le monde" });getAlternativeTranslationsThe other ways Google would have translated the same thing.Arguments
Name Type Required textstringYes tostringYes fromstringNo Returns
GoogleTranslateAlternativeTranslationsinterface GoogleTranslateAlternative { text: string; backends: number[]; } interface GoogleTranslateAlternativeSegment { sourceSegment: string; offsets: { begin: number; end: number }; alternatives: GoogleTranslateAlternative[]; } interface GoogleTranslateAlternativeTranslations { text: string; targetLanguage: string; sourceLanguage: string; segments: GoogleTranslateAlternativeSegment[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/getAlternativeTranslationsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.getAlternativeTranslations({ text: "The meeting was moved to Friday.", to: "French" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.getAlternativeTranslations({"text": "The meeting was moved to Friday.", "to": "French"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/getAlternativeTranslations \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"text":"The meeting was moved to Friday.","to":"French"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.getAlternativeTranslations({ text: "The meeting was moved to Friday.", to: "French" });getDefinitionsWhat a word MEANS, in its own language.Arguments
Name Type Required wordstringYes languagestringYes Returns
GoogleTranslateWordDefinitionsinterface GoogleTranslateDefinitionExample { text: string; } interface GoogleTranslateDefinition { definitionId: string; gloss: string; examples: GoogleTranslateDefinitionExample[]; subject?: string[]; register?: string[]; } interface GoogleTranslateDefinitionSense { partOfSpeech: string; definitions: GoogleTranslateDefinition[]; } interface GoogleTranslateWordDefinitions { word: string; language: string; senses: GoogleTranslateDefinitionSense[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/getDefinitionsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.getDefinitions({ word: "run", language: "English" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.getDefinitions({"word": "run", "language": "English"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/getDefinitions \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"word":"run","language":"English"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.getDefinitions({ word: "run", language: "English" });getSynonymsOther words that mean the same thing, grouped by sense rather than thrown into one list, and labelled where Google knows the register.Arguments
Name Type Required wordstringYes languagestringYes Returns
GoogleTranslateWordSynonymsinterface GoogleTranslateSynonymGroup { definitionId: string; synonyms: string[]; register?: string[]; } interface GoogleTranslateSynonymSense { partOfSpeech: string; groups: GoogleTranslateSynonymGroup[]; } interface GoogleTranslateWordSynonyms { word: string; language: string; senses: GoogleTranslateSynonymSense[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/getSynonymsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.getSynonyms({ word: "run", language: "English" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.getSynonyms({"word": "run", "language": "English"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/getSynonyms \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"word":"run","language":"English"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.getSynonyms({ word: "run", language: "English" });listHistoryWhat the signed-in person has translated recently, newest first.Arguments
None.
Returns
neverHTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/listHistoryCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.listHistory();import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.listHistory())curl -X POST https://api.bowmark.ai/v1/sites/google_translate/listHistory \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.listHistory();listLanguagesEvery language this site supports.Arguments
Name Type Required hlstringNo Returns
GoogleTranslateLanguage[]interface GoogleTranslateLanguage { code: string; name: string; sourceSupported: boolean; targetSupported: boolean; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/listLanguagesCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.listLanguages();import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.listLanguages())curl -X POST https://api.bowmark.ai/v1/sites/google_translate/listLanguages \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.listLanguages();listSavedPhrasesThe phrases the signed-in person starred.Arguments
None.
Returns
neverHTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/listSavedPhrasesCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.listSavedPhrases();import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.listSavedPhrases())curl -X POST https://api.bowmark.ai/v1/sites/google_translate/listSavedPhrases \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.listSavedPhrases();lookupWordThe full "translations of <word>" panel for one word.Arguments
Name Type Required wordstringYes tostringYes fromstringYes hlstringNo Returns
GoogleTranslateWordLookupinterface GoogleTranslateWordCandidate { word: string; reverseTranslations: string[]; score: number; } interface GoogleTranslateWordSense { partOfSpeech: string; candidates: GoogleTranslateWordCandidate[]; } interface GoogleTranslateWordLookup { word: string; targetLanguage: string; sourceLanguage: string; senses: GoogleTranslateWordSense[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/lookupWordCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.lookupWord({ word: "run", to: "Spanish", from: "English" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.lookupWord({"word": "run", "to": "Spanish", "from": "English"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/lookupWord \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"word":"run","to":"Spanish","from":"English"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.lookupWord({ word: "run", to: "Spanish", from: "English" });romanizeA Latin-alphabet (or phonetic) rendering of the translation, or the source, printed under a translation into a non-Latin script.Arguments
Name Type Required textstringYes tostringYes fromstringNo Returns
GoogleTranslateRomanizationinterface GoogleTranslateRomanization { text: string; targetLanguage: string; sourceLanguage: string; detected: boolean; targetRomanization?: string; sourceRomanization?: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/romanizeCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.romanize({ text: "Good morning, how are you?", to: "Japanese" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.romanize({"text": "Good morning, how are you?", "to": "Japanese"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/romanize \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"text":"Good morning, how are you?","to":"Japanese"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.romanize({ text: "Good morning, how are you?", to: "Japanese" });speakHear args.text spoken in args.language, as the MP3 the site's own speaker button plays.Arguments
Name Type Required textstringYes languagestringYes slowbooleanNo Returns
GoogleTranslateSpeechinterface GoogleTranslateSpeech { audioBase64: string; contentType: string; chunkCount: number; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/speakCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.speak({ text: "Hello", language: "Spanish" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.speak({"text": "Hello", "language": "Spanish"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/speak \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"text":"Hello","language":"Spanish"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.speak({ text: "Hello", language: "Spanish" });translateTurn text into another language.Arguments
Name Type Required textstring | readonly string[]Yes tostringYes fromstringNo model"classic" | "advanced"No Returns
GoogleTranslateResult[]interface GoogleTranslateResult { source: string; translated: string; targetLanguage: string; sourceLanguage: string; detected: boolean; genderAlternatives?: { gender: string; translated: string }[]; model?: "classic" | "advanced"; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/translateCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.translate({ text: "hello", to: "French" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.translate({"text": "hello", "to": "French"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/translate \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"text":"hello","to":"French"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.translate({ text: "hello", to: "French" });translateDocumentTranslate a whole document.Arguments
Name Type Required fileBase64stringYes mimeTypestringYes tostringYes fromstringNo Returns
GoogleTranslateDocumentResultinterface GoogleTranslateDocumentResult { translatedBase64: string; mimeType: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/translateDocumentCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.translateDocument({ fileBase64: "<fileBase64>", mimeType: "<mimeType>", to: "<to>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.translateDocument({ "fileBase64": "<fileBase64>", "mimeType": "<mimeType>", "to": "<to>" }))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/translateDocument \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"fileBase64":"<fileBase64>","mimeType":"<mimeType>","to":"<to>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.translateDocument({ fileBase64: "<fileBase64>", mimeType: "<mimeType>", to: "<to>" });translateImageRead the text in a picture and translate it.Arguments
Name Type Required imageBase64stringYes mimeTypestringYes tostringYes fromstringNo Returns
GoogleTranslateImageResultinterface GoogleTranslateImageResult { translatedImageBase64: string; mimeType: string; sourceText: string; translatedText: string; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/translateImageCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.translateImage({ imageBase64: "<imageBase64>", mimeType: "<mimeType>", to: "<to>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.translateImage({ "imageBase64": "<imageBase64>", "mimeType": "<mimeType>", "to": "<to>" }))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/translateImage \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"imageBase64":"<imageBase64>","mimeType":"<mimeType>","to":"<to>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.translateImage({ imageBase64: "<imageBase64>", mimeType: "<mimeType>", to: "<to>" });translateWebPageRead somebody else's web page in your own language.Arguments
Name Type Required urlstringYes tostringYes fromstringNo Returns
GoogleTranslateWebPageinterface GoogleTranslateWebPageSegment { original: string; translated: string; } interface GoogleTranslateWebPage { url: string; title: string; translatedTitle: string; targetLanguage: string; sourceLanguage: string; detected: boolean; segments: GoogleTranslateWebPageSegment[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/translateWebPageCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.translateWebPage({ url: "<url>", to: "<to>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.translateWebPage({"url": "<url>", "to": "<to>"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/translateWebPage \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"url":"<url>","to":"<to>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.translateWebPage({ url: "<url>", to: "<to>" });
deleteHistoryDelete from the signed-in person's translation history.Arguments
Name Type Required textstringYes Returns
neverHTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/deleteHistoryCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.deleteHistory({ text: "<text>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.deleteHistory({"text": "<text>"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/deleteHistory \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"text":"<text>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.deleteHistory({ text: "<text>" });savePhraseStar a translation so it joins the signed-in person's saved phrases.Arguments
Name Type Required textstringYes fromstringYes tostringYes Returns
neverHTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/savePhraseCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.savePhrase({ text: "<text>", from: "<from>", to: "<to>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.savePhrase({"text": "<text>", "from": "<from>", "to": "<to>"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/savePhrase \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"text":"<text>","from":"<from>","to":"<to>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.savePhrase({ text: "<text>", from: "<from>", to: "<to>" });unsavePhraseTake a phrase back out of the signed-in person's saved phrases.Arguments
Name Type Required textstringYes fromstringYes tostringYes Returns
neverHTTP endpoint
POST https://api.bowmark.ai/v1/sites/google_translate/unsavePhraseCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.google_translate.unsavePhrase({ text: "<text>", from: "<from>", to: "<to>" });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.google_translate.unsavePhrase({"text": "<text>", "from": "<from>", "to": "<to>"}))curl -X POST https://api.bowmark.ai/v1/sites/google_translate/unsavePhrase \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"text":"<text>","from":"<from>","to":"<to>"}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.google_translate.unsavePhrase({ text: "<text>", from: "<from>", to: "<to>" });
Use it from code
- Node
npm i @bowmark/web - Python
pip install bowmark-web - HTTP
POST /v1/sites/google_translate/<function> - OpenAPI
/v1/sites/google_translate/openapi.json