Jake Levine

Muse AI Connectors COULD be the App Store moment for AI, and I think BILLIONS of dollars are up for grabs.

youtube.com/watch?v=84q4WA…

When Apple opened the App Store in 2008, it gave developers access to millions of iPhone users, and less than two years later Apple had paid them $1 billion.

Muse could create a similar opening, except instead of building apps people have to find and open, founders can build services a personal agent calls while completing a task.

In today's episode of @startupideaspod, I explain how Muse Connectors work, how to build one with Codex or Claude Code, how people could discover them and 4 startup ideas I’d investigate right now.

Who knows whether Muse reaches 10 million, 100 million or one billion users, but whether Muse wins or not, personal agents are becoming a new place where demand gets routed and money changes hands.

So, needless to say, it's a really good time to be experimenting with Muse AI connectors.

Nobody knew which iPhone apps would win in 2008, but the developers who started experimenting early learned what the rest of the market wouldn’t understand for years.

This feels like another one of those moments.

We're excited to partner with @Meta to enable PayPal customers to seamlessly shop and check out using their @Muse personal AI agents across PayPal merchants worldwide.

Here at Expedia, we love planning trips. Soon, your personal AI agent will too. We're joining @Muse: tell it where you're headed and it can work with Expedia to sort your hotels, and everything in between.

We are excited to announce we are partnering deeply with Muse to enable agentic checkout with Shop Pay on all Shopify stores, offering people an easy and delightful way to shop and check out with Muse.

Opening access for developers to build Muse connectors. You bring the API -- Muse brings the agent, the browser, and the context of what the person actually wants. People reach your service just by asking for it, and their agent takes it from there.

New connectors are live today. Come build with us. muse.ai/platform

Get more done with Muse with two new connectors:
✅ @meetgranola: your meeting notes
✅ @NotionHQ: your docs
…and more coming soon!

Ask Muse or connect in settings.

muse for mac is HERE!! your agent can now get stuff done right on your computer — files, messages, calendar, notes, all of it.

you're in control of what it can access, and it always asks before doing anything sensitive.

Add it to the dock - ai.meta.com/muse/download/

You can bring your @Muse avatar as your personalized app icon by long pressing on the Muse app icon and changing the view to a widget

One thing I really like about Muse: I trust its Secrets Store

Every time you use an agentic app, you have to make a call. Do you trust how it stores and uses secrets? Passwords, keys, credentials, etc.

In the early days of vibe coding, secrets were often stored … literally in plain text by agents. And leaked all across the web. Not great.

Today, I just … don’t know. Will some hot startup have a security breach and leak my secrets? Of course. It still happens all the time. Every day. You almost have to assume any startup you give your credentials and secrets to will have a breach. And agents just need … a lot of them to work well. A lot of secrets.

Maybe it’s ironic I trust “Facebook” more than 99% of apps with my secrets. But the app seems to have made it part of its core from Day 1, to do it right

I mostly trust Muse (for now at least) with my credentials. I’m making a gut call based on experience, but I think they’ve truly put in the work. In the AI Age, I mostly don’t truly most other apps with them.

For that reason alone, I’ll probably end use doing more with Muse than other general agents.

2/ also for the first time ever, through this partnership, we are providing agentic payment protection – if Muse makes a mistake, a refund guarantee will cover it. More details here: support.link.com/questions/what…

I sent Muse a blog post I wrote about losing a sentimental pair of pants that I bought in Hong Kong 8 years ago…

3 days later, it found the exact same pair on eBay, in my size 🥹

We’re testing a solution for this as we speak…