기본 콘텐츠로 건너뛰기

Google Opal Shuts Down Nov 17: Back Up and Move Workflows

Summary: Google is turning off Opal, its Labs experiment for building AI mini-apps, on November 17, 2026, together with "Gems made by Labs" in Gemini. Your workflow files stay in Google Drive under My Drive > Opal, but they will stop running, and they will not be converted for you. This guide shows what to save before the deadline, how to rebuild an Opal workflow as a skill in Gemini or a prompt in Google AI Studio, and where the two replacements fall short.

Everything below was checked on Google's own pages today: the Opal FAQ on developers.google.com, the notice on opal.google, and the Gemini Apps Help articles about skills and the Gems-to-skills transition. I did not have Opal workflows of my own to migrate for this post, so the steps are taken from the official help pages rather than from a hands-on migration log. Where Google's pages are silent (for example, the exact hour on November 17 when the service stops), I say so instead of guessing.

What is actually shutting down, and when

The Opal FAQ, under "Why is Opal ("Gems made by Labs") graduating, and where can I build next?", says the experimental Opal service at opal.google.com and "Gems made by Labs" in Gemini will turn off on November 17, 2026. The banner on the opal.google landing page repeats the date and adds that active workflows will no longer run after it. Google describes this as a "graduation": what the team learned from Opal fed into skills in Gemini, which Google calls its newest core feature for saving and reusing custom instructions in chat.

Three separate things change on that date, and it helps to keep them apart:

  • Opal mini-apps (built at opal.google or as "My Gems from Labs" in the Gemini web app) stop running. The files remain in your Drive.
  • Classic Gems on personal accounts for people 18 or over become read-only on November 17, 2026 and are removed from Gemini Apps on December 17, 2026. Unlike Opal, these are converted to skills automatically.
  • The open-source code: the Breadboard and Opal ADK repositories on GitHub (breadboard-ai/breadboard and breadboard-ai/opal-adk) move to read-only, archived mode. They stay public under their licenses, so you can still clone or fork them, but hosted endpoints such as breadboard-ai.web.app turn off alongside opal.google.com.

Work and school accounts follow a later Gems timeline (March 2027 for Workspace business, enterprise and non-profit accounts; June 2027 for education), but that schedule is for Gems. The Opal FAQ gives a single date for Opal itself.

The one sentence most people will miss

The FAQ is blunt: "Existing Opal workflows will not automatically migrate, so you'll need to set up any new skills or prompts manually." If you read headlines about Gems turning into skills automatically, that applies to classic Gems, not to Opal mini-apps or Gems from Labs. Nothing is lost from Drive, but nobody is going to rebuild your workflow for you.

Step 1: Back up your Opal workflows before November 17

Copy prompts while the editor still works

Google's own recommendation for the period before November 17 is to open each workflow on opal.google.com and copy any prompts or custom instructions you want to keep. Do this now, while you can still see the graph. A practical routine for each mini-app:

  1. Open the workflow in the Opal editor (the FAQ notes the editor view is optimized for a desktop computer).
  2. Go through the steps in order and paste each prompt into a plain text document. Note what each step takes as input and what it produces, because that ordering is the part you will have to rebuild by hand.
  3. Write down which steps used tools such as web search, and which produced images, video or audio. Those are the steps that will need the most rethinking later.
  4. Run the app once with a typical input and save the output. A known-good example makes it much easier to check whether the rebuilt version behaves the same.

Download the files from My Drive > Opal

The FAQ says your saved workflow files live in your personal Google Drive in a folder called "Opal" (My Drive > Opal), and that Google will not delete or touch them when the service turns off. Even so, a local copy costs nothing. Google Drive Help describes the standard download: go to drive.google.com, click a file, Ctrl-click (Windows) or Command-click (Mac) any others, then right-click and choose Download.

Reading the files after the shutdown

After November 17 you can still get your prompt text by downloading the workflow files from that Drive folder and opening them in any plain text editor, such as Notepad or TextEdit. That is the official fallback. It means the prompts survive, but the clickable app does not. Expect to read raw text rather than a tidy step-by-step view, which is another reason to make your own clean notes before the deadline.

If you shared an Opal

The FAQ explains that sharing an Opal also shares the underlying Drive file. Once workflows no longer run, a shared link stops being a working app, so tell anyone who relies on one of your mini-apps where its replacement will live. If you shared with editor view enabled, remember that those people can see your prompts through the Drive file; review sharing on that folder if any prompts are sensitive.

Step 2: Choose where each workflow should go

Google points to two destinations. Skills in Gemini are meant for everyday tasks and customized chat prompts. Google AI Studio is meant for deeper prototyping and custom prompt experiments. A rough rule from the help pages: if your mini-app was mostly "follow these instructions with this reference material", it maps well to a skill. If it chained several model calls, generated media, or you want to turn it into a real app, AI Studio and the Gemini API are the closer fit.

Option Cost / free access Limits verified today Good for
Opal (until Nov 17, 2026) No price is listed in the Opal FAQ Stops running on Nov 17, 2026; files stay in My Drive > Opal; no automatic migration Chaining text, image and video prompts with tools like web search into mini-apps
Skills in Gemini Available without a Google AI subscription Must be 18+ and signed in; up to 100 MB of files per skill; unlimited skills but only 100 active at once; scripts can't make internet requests; doesn't yet work with Canvas, Deep Research, Create video or Create music Reusable instructions plus reference files, called with / (or @) in any chat or picked up automatically
Google AI Studio (Gemini API) Free tier with free input and output tokens; paid tier for production Free tier has limited access to certain models and content is used to improve Google's products; paid tier adds higher rate limits, context caching and the Batch API at 50% lower cost Prototyping multi-step prompts and building apps on Gemini, image, video and speech models

Step 3: Rebuild a workflow as a skill in Gemini

The Gemini Apps Help article "Create & manage skills for Gemini Apps" lists several ways to create a skill. For an Opal migration, the manual route gives you the most control:

  1. Go to gemini.google.com, click Customize in the sidebar (under More if you don't see it), then Skills at the top.
  2. Click Create then Create manually.
  3. Enter a name, a description of what the skill does, and the instructions. This is where your copied Opal prompts go. Merge the steps into one ordered set of instructions ("First do X with the input, then use that result to do Y").
  4. Optionally click Add to attach reference files from your device, Google Drive, Google Photos or Notebooks, then click Create.

If you prefer working in files, you can write a SKILL.md yourself and use Create then Upload. You can select a single SKILL.md, or a folder or .zip with SKILL.md in its main folder. The skill name inside SKILL.md must be lowercase with hyphens, such as lecture-quiz-maker, and all uploaded files together must stay under 100 MB. Help also notes that you can import skills made on other platforms by uploading their SKILL.md files, which makes a SKILL.md a handy portable backup of your Opal logic.

Once it exists, type / (or @, if you have the updated shortcut) in any chat and pick the skill, or let Gemini apply it automatically when it judges the skill relevant. You can also reference one skill from another to build longer chains, which is the nearest thing to Opal's multi-step graph inside a chat.

Step 4: Rebuild it in Google AI Studio instead

For mini-apps that relied on media generation or several model calls, AI Studio is the better landing spot. Its welcome page offers Gemini, Nano Banana for images, Veo for video, Gemini TTS and Lyria for music, plus an app gallery with projects you can remix. Paste your copied prompts into a new prompt, test each former Opal step one at a time, and keep the sample output from your backup to compare against. When the chain works, get an API key in AI Studio and move it into code with the official SDKs.

Two practical points from the pricing and rate-limit docs: on the free tier, your content is used to improve Google's products, and rate limits apply per project rather than per API key, with daily request quotas resetting at midnight Pacific time. Linking a billing account moves a project to Tier 1, which also brings a spend-based limit of $10 per rolling 10 minutes.

Editor's note

Who this is for: anyone with mini-apps in My Drive > Opal or under "My Gems from Labs", and anyone who shared an Opal with a team or class. Who can skip it: people who only used classic Gems; those convert to skills automatically, though you should still download any knowledge files that won't carry over.

Gotchas and fixes:

  • "My Opal disappeared from Drive": it shouldn't. Google says it won't delete or touch the files. Check My Drive > Opal while signed in to the same personal account you used for Opal.
  • Skill upload fails: look for hidden binary files like .DS_Store or .pyc in the folder, make sure SKILL.md sits in the root folder, and check the name is lowercase-with-hyphens. Legacy .doc/.xls/.ppt files, audio, video and archives are not supported.
  • Skill can't fetch a web page: skill scripts can't perform requests to external websites. Move that step to AI Studio or the API.
  • Image or video step has nowhere to go: Create video and Create music don't work with skills yet, so rebuild those steps in AI Studio.
  • Exact cutoff time: Google gives a date, not a time zone. Finish your backup by November 16 in your own time zone to be safe.
  • 429 RESOURCE_EXHAUSTED in AI Studio: you hit a rate or spend limit. Wait and retry, shorten prompts or outputs, or request a higher limit.

FAQ

Will Google delete my Opal files?

No. According to the Opal FAQ, workflow files stay in My Drive > Opal and Google won't delete or touch them when the service turns off. They just stop running as apps after November 17, 2026.

Will my Opal mini-apps turn into skills automatically, like Gems?

No. Classic Gems are transitioned to skills automatically, but the Opal FAQ says existing Opal workflows will not migrate on their own. You need to recreate them as skills or prompts yourself.

Do I need a paid Google AI plan to use skills?

No. Gemini Apps Help says skills are available without a Google AI subscription and can be used in your chats. You do need to be 18 or over and signed in.

Related reading on this blog: NotebookLM Playbook: PDFs to Briefs and Audio Overviews if your Opal apps turned documents into summaries or audio, n8n Playbook: Draft Inbox Replies With an AI Agent for workflows that need real automation outside a chat, and Antigravity CLI Setup: Move From Gemini CLI Step by Step for another Google tool migration.

Copy your prompts this week while the editor still opens; rebuilding is easy, but rediscovering lost prompt logic is not.

See the official screenshots at the end of this post. Image credit: my own screenshots of the public opal.google page, the Opal FAQ on developers.google.com and Gemini Apps Help (support.google.com).

Last verified: 2026-11-11
Sources: Opal (opal.google), Opal FAQ, About the transition from Gems to skills, Create & manage skills for Gemini Apps, Gems from Google Labs help, Google Drive: Download a file, Google AI Studio, Gemini API pricing, Gemini API rate limits.

Uploaded Image Uploaded Image Uploaded Image Uploaded Image

댓글

이 블로그의 인기 게시물

브루(Vrew) 요금제 정리: 무료 200크레딧으로 어디까지 쓸 수 있나

요약: 영상 자막 편집기 브루(Vrew)는 2026년 4월 22일부터 기능별 사용량 대신 '통합 크레딧' 하나로 요금을 매겨요. 무료 플랜은 매달 200크레딧이 들어오는데, 음성 분석(자동 자막)만 쓰면 최대 200분 분량이라 짧은 영상 몇 편은 충분하지만, 영상 초반의 Vrew 마크를 지우려면 유료 플랜이 필요해요. 이 글은 브루 공식 가격 페이지와 고객센터 공지를 2026년 10월 23일에 다시 확인해, 크레딧 계산법과 플랜 고르는 기준을 정리했어요. 목차 브루 요금 구조: 통합 크레딧이란 1크레딧으로 할 수 있는 일 무료 플랜으로 되는 것과 안 되는 것 유료 플랜 4종 비교표 월 구독·연 구독·기간 이용권, 무엇이 이득일까 상황별 추천: 학생·직장인·유튜버 크레딧이 모자랄 때: 사용 기록 지우개 업그레이드·환불 전에 알아둘 규칙 운영자 메모, 자주 묻는 질문 1. 브루 요금 구조: 통합 크레딧이란 예전 브루는 플랜마다 음성 분석 몇 분, AI 목소리 몇 자, 번역 몇 자처럼 기능별 한도를 따로 줬어요. 그런데 보이저엑스(브루 운영사)가 실제 이용 데이터를 보니 한두 가지 기능만 집중해서 쓰는 사람이 훨씬 많았다고 해요. 그래서 2026년 4월 22일 업데이트부터는 플랜마다 정해진 크레딧을 받고, 어떤 기능에 쓸지는 내가 정하는 방식으로 바뀌었어요. 플랜별 매달 지급량은 Free 200 · Light 2,000 · Standard 10,000 · Business 12,000크레딧 이에요. 공식 공지에 따르면 무료 사용자는 매달 1일에, 유료 사용자는 매월 결제일에 충전되고, 남은 크레딧은 다음 달로 넘어가지 않아요. 바뀐 뒤 무엇이 좋아졌나 공지에 나온 예시가 이해하기 쉬워요. Standard 플랜에서 AI 목소리만 쓰면 예전에는 최대 50만 자였지만, 지금은 1만 크레딧을 모두 AI 목소리에 몰아 쓸 수 있어 최대 100만 자가 돼요. 반대로 여러 기능을 골고루 쓰는 사람은 크레딧이 기능끼리 ...

에이닷 전화 vs 익시오: 통화 녹음·AI 요약, 내 통신사·폰에 맞는 앱 고르기

요약: 통화 녹음과 AI 요약을 쓰고 싶다면 먼저 내 통신사와 휴대폰 종류부터 확인해야 해요. LG유플러스 고객은 무료 앱 익시오, SKT 고객은 에이닷 전화가 가장 기능이 많고, 다른 통신사의 아이폰 사용자는 iOS에 들어 있는 기본 통화 녹음·전사 기능이 현실적인 선택이에요. 이 글은 2026-10-30에 SKT·LG유플러스·Apple 공식 페이지와 도움말을 직접 확인해 세 가지를 비교했어요. 왜 '앱 기능'보다 '내 조건'을 먼저 봐야 할까요 통화 요약 앱 비교 글을 보면 기능 목록부터 나열하는 경우가 많아요. 그런데 실제로 써 보려는 사람에게 가장 먼저 걸리는 건 기능이 아니라 이용 조건 이에요. 익시오는 LG유플러스 고객 전용이고, 에이닷 전화는 누구나 설치할 수 있지만 통역콜·baro 같은 기능은 SKT 고객에게만 열려 있어요. 아이폰은 운영체제 구조상 앱이 통화를 마음대로 녹음하기 어려워서, 통신사 앱이 안 되면 iOS 기본 기능에 기대야 해요. 그래서 이 글은 "어떤 앱이 제일 좋은가"보다 "내 통신사·내 폰에서 무엇을 쓸 수 있나" 순서로 정리했어요. 기능 차이는 그다음이에요. 한눈에 비교: 에이닷 전화 vs 익시오 vs 아이폰 기본 통화 녹음 구분 에이닷 전화 (SKT) 익시오 ixi-O (LG유플러스) 아이폰 기본 통화 녹음 (Apple) 누가 쓸 수 있나 통신사 구분 없이 앱 설치 가능. 단, 통역콜·baro·그룹통화 발신 등은 SKT 고객 전용 U+ 모바일 고객. 타 통신사, 만 14세 미만, 선불요금제 고객은 제한 (공식 사이트 FAQ상 알뜰폰 중 KB리브엠은 예외) 지원 국가·언어의 iPhone 사용자. 1명과 통화할 때만 녹음 가능 지원 OS·기기 Android 10 이상, iOS 16 이상 iOS 17 이상 iPhone 12 이상 / Android...

타입캐스트 vs 클로바더빙: 수익 나는 채널에 쓸 때 무료·유료 조건 비교

요약 타입캐스트와 클로바더빙은 둘 다 글만 입력하면 한국어 AI 목소리를 만들어 주는 서비스지만, 무료로 쓸 수 있는 범위가 꽤 다릅니다. 클로바더빙 무료는 매달 15,000자까지 쓸 수 있는 대신 수익이 생기는 채널에는 쓸 수 없고, 타입캐스트 무료는 처음에 주는 약 5분 분량의 다운로드로 끝나는 체험판에 가깝습니다. 광고 수익이 붙은 유튜브나 쇼핑몰 영상에 넣을 목소리라면 처음부터 유료 플랜 조건을 보고 고르는 편이 안전합니다. 왜 이 두 서비스를 같이 비교하나 쇼츠 내레이션, 수업 자료 더빙, 상품 소개 영상처럼 "내 목소리 대신 AI 목소리"가 필요할 때 한국 사용자들이 가장 많이 찾는 곳이 네오사피엔스의 타입캐스트 와 네이버의 클로바더빙 입니다. 기능 소개만 보면 비슷해 보이지만, 실제로 문제가 되는 건 음질보다 "이 목소리를 내 채널에 써도 되나" 입니다. 그래서 이 글은 기능 나열보다 무료 사용 범위, 출처 표기 의무, 유료 플랜 가격과 다운로드 한도를 중심으로 정리했습니다. 모든 수치는 2026년 10월 25일에 두 서비스의 공식 요금 페이지와 고객센터 도움말에서 직접 다시 확인한 내용입니다. 한눈에 보는 비교표 항목 타입캐스트 무료 타입캐스트 베이직 클로바더빙 Free 클로바더빙 Standard 월 요금 0원 9,900원 (연간 결제 시 10% 할인) 0원 19,900원 (네이버 클라우드 플랫폼에서 결제) 무료·포함 사용량 음성 생성·재생 무제한, 다운로드는 최초 3,000크레딧(약 5분) 1회 지급 월 30,000크레딧(약 60분) 월 15,000자, 다운로드 월 20회, 프로젝트 누적 5개 (매월 1일 갱신) 월 30,000자, 다운로드 월 40회, 프로젝트 누적 20개 쓸 수 있...