I added Gemini to Alexa+ — and it unlocked a whole new level of customization

Features
By published

This is my hack for giving Alexa+ a Google Gemini boost

Alexa+ on Echo Show
(Image credit: Future)

I have used Alexa+ for months and absolutely love how it keeps my chaotic family of five organized, scheduled and entertained. But as an AI editor eager to test the limits of AI, I did something a little nerdy: I added Google Gemini to Alexa+.

Let me be clear, this is not some kind of illegal jailbreak or backend tech hack. Honestly, I wouldn’t change what Amazon has done with its updated assistant, not just because it’s impossible to do, but because I genuinely appreciate the assistant.

But by creating a custom Alexa skill that quietly hands certain questions off to Gemini instead, Alexa is even more informative, almost like a second personality. And it turns out, this is one of the easiest ways to get way more out of Alexa+ right now.

You can’t swap Alexa+ — but you can extend it

There’s no official way to completely swap Alexa+’s underlying AI with Gemini, ChatGPT or anything else. Amazon controls the core assistant — and honestly, that’s fine. Alexa+ supports home automation in numerous ways and is good at:

  • smart home control
  • reminders and routines
  • quick answers
  • family logistics

But Alexa has always been modular. And that’s where custom skills come in. With a custom skill, Alexa can:

  • send a request to your own backend
  • process it however you want
  • speak the response like it’s native

In my setup, Alexa hands certain questions to Gemini, then reads Gemini’s response aloud. With this custom skill or hack, the only difference is the invocation:

“Alexa, open Gemini Helper.”

From that point on, Alexa becomes Gemini-powered — at least for that conversation.

Why Gemini works so well inside Alexa

I’ve tested nearly every major chatbot, and Gemini is particularly well suited for voice-based use:

  • Stronger reasoning for open-ended questions
  • Better follow-ups, which matter when you’re talking
  • More flexible tone, especially for creative or exploratory prompts

Now, when I ask Alexa+ for a fact, it’s fast and reliable. Having Gemini-powered Alexa is also great for brainstorming help — from meals to story ideas, deeper explanations (for the kids' homework or my own research), comparisons or “talk this through with me” answers.

Gemini consistently gives better results. Sure, I could use Gemini Live on my phone, but with Alexa+ right there on my countertop, it just makes sense. Once the skill is live, Alexa becomes a very different kind of assistant. It’s a level of flexibility Alexa simply doesn’t offer natively — at least not yet.

How to add Gemini to Alexa+

You don’t need special access, insider permissions or developer experience. I'm a pop science nerd, not a developer, so if I can do it, you can do it. All you need is an Amazon Account and a Google account along with a device that runs Alexa+ (Echo and Echo Dot models from 2nd generation onward).

Step 1: Get a Gemini API key

You’ll need access to Gemini’s API, which Google provides through Google AI Studio. Sign in with your Google account. Create an API Key, Make sure Gemini is enabled (e.g., gemini-pro) is enabled. This key lets your Alexa skill send questions to Gemini.

Step 2. Create a Custom Alexa Skill

You do this through the Alexa Developer Console. Simply choose: "Custom Skill" and Alexa-hosted. Choose Node.js or Python as the backend. This creates a skill that runs in Amazon’s cloud but can call outside services like Gemini.

Step 3. Set an invocation name

This is what you say to Alexa to activate the skill. In my case, I say, "Alexa, open Gemini Helper." It's important to note that Amazon’s naming rules apply (no trademarked terms, must sound natural).

Step 4. Add one intent using AMAZON.SearchQuery

Now, instead of defining dozens of intents, you can use Amazon’s free-form query intent, which captures anything the user says after opening the skill. This lets Alexa pass the full spoken question to Gemini without heavy intent design.

Step 5. Route the question to Gemini’s "generateContent" API

This screen is misleading, but stay with me. Scroll down to "General," click the arrow and scroll until you see "+ add intent" to create a custom intent.

Select it. Important: do not pick: local search, books, music, weather etc. Those are domain-specific and will break the Gemini handoff.

From there, click enter and return the response as Alexa speech. You'll know it's working when it works in the Alexa Test Console. If it works in the Alexa Test Console, it will work on your Echo devices automatically.

FAQ

I've been getting several emails from users who have seen success and a few who have gotten tripped up. For that reason, here are a few Q&A that may help clear things up. Feel free to reach out if you still experience trouble. This works so beautifully — and is totally worth the extra time to get it right.

  • Do I need Alexa+ to do this? Yes. This setup relies on Alexa+ features that allow more flexible intent handling and backend responses. It won’t work on legacy Alexa skills alone.
  • I don’t see an intent to select — what am I missing? That’s expected. The intent doesn’t exist by default. Under General, click + Add Intent and create a custom intent (for example, AskGeminiIntent). Once you create it, that is the intent you select.
  • What exactly does the custom intent do? The custom intent captures what is said to Alexa, passes that text to your backend and lets your backend forward the request to Gemini. Then, it returns Gemini’s response back to Alexa
  • Where do I paste the Gemini API key? You don’t paste the Gemini API key into the Alexa intent screen. The key goes in your backend code (either Alexa-hosted code or an AWS Lambda function), ideally as an environment variable (for example: GEMINI_API_KEY). The intent only captures what you say — the backend is what actually sends the request to Google Gemini.
  • Why can’t I use Local Search, Weather, Books, or Music intents? Those are domain-specific built-in intents tied to Alexa’s own services. Selecting them will override your custom logic and break the Gemini handoff. You must use a custom intent for this to work.
  • Do I need to write code for this to work? Yes, minimally. You need backend code that: listens for your custom intent, reads your spoken text and sends it to the Gemini API, then returns the response to Alexa. Struggling to write the code? Open the Gemini app or browser and type this prompt:

    Act as a senior Alexa skill engineer. I need minimal, production-safe Node.js code for an Alexa skill using ASK SDK v2. Requirements:
    • Custom intent name: AskGeminiIntent
    • Slot name: query (AMAZON.SearchQuery)
    • Call Google Gemini using the generateContent endpoint
    • Read API key from process.env.GEMINI_API_KEY
    • Handle missing slots and API errors gracefully
    • Keep spoken responses under ~30 seconds
    • Output only the code for index.js
    • Optimize for voice assistance
  • I used your prompt and Gemini gave me the code, now what? You don’t paste this code into the Alexa intent screen — you paste it into the Code section of your Alexa skill (Alexa-hosted or Lambda).
  • What’s the most common reason this fails? The most commen reason is your intent name in the Alexa UI doesn’t exactly match the intent handler name in the backend. You also might run into issues if a built-in Alexa intent was selected and not your custom one. Or, lastly, the Gemini API key isn’t set correctly as an environment variable
  • Is this officially supported by Amazon or Google? No. But it's not a jailbreak or illegal either. This is a custom integration using public APIs. It works reliably, but it’s not a one-click, officially endorsed feature — which is why setup matters.
  • Is this safe to use with my personal data? Your data flows through your own backend. As always, you should keep API keys in environment variables, avoid logging sensitive requests, review Google’s Gemini API data usage policies
  • What can I use this for once it’s working? Deeper reasoning than Alexa alone provides summarizing long text or ideas, planning trips or schedules, rewriting or brainstorming out loud. For example, I brainstorm with Gemini via Alexa+ while I'm cooking dinner.
  • Is this replacing Alexa’s intelligence? Nope — it’s augmenting it.
  • This still isn't working?! Even if every step is correct, some readers may still hit friction because: Amazon’s console UI varies by region/account, Node version differences (fetch availability) can break code, some Alexa+ experiences behave differently across devices. This will work if you’re using a custom intent + backend code + valid Gemini key. If any one of those is missing, it won’t.

Bottom line

This looks like a lot of steps, but in practice it’s straightforward. This cusom skill gives Alexa+ a Gemini boost while keeping Alexa+ fast and practical. With this hack, Gemini steps in when you want depth, context or real reasoning — and that balance is what makes this setup worth it.

Custom skills are there to tailor Alexa+ to you and your needs, and the result is an assistant that thinks and reasons before it speaks. Give it a try and let me know what you think in the comments.

Google News

Follow Tom's Guide on Google News and add us as a preferred source to get our up-to-date news, analysis, and reviews in your feeds.

More from Tom's Guide

Amanda Caswell
Amanda Caswell
AI Editor

Amanda Caswell is an award-winning journalist, bestselling YA author, and one of today’s leading voices in AI and technology. A celebrated contributor to various news outlets, her sharp insights and relatable storytelling have earned her a loyal readership. Amanda’s work has been recognized with prestigious honors, including outstanding contribution to media.

Known for her ability to bring clarity to even the most complex topics, Amanda seamlessly blends innovation and creativity, inspiring readers to embrace the power of AI and emerging technologies. As a certified prompt engineer, she continues to push the boundaries of how humans and AI can work together.

Beyond her journalism career, Amanda is a long-distance runner and mom of three. She lives in New Jersey.

You must confirm your public display name before commenting

Please logout and then login again, you will then be prompted to enter your display name.