Meta Muse Review: A Personal AI Agent for Family Life

It's the weekend, the kids are out of the house, and that means it’s time to test a new AI agent. This time, I dove into Muse, Meta's new personal AI assistant. Yes, that Meta. They're positioning it as your personal agent that gets things done, and after spending a few hours with it, I have some thoughts.

Looking at its landing page, it was immediately clear who Muse is for: me. It's for millennials with kids, maybe even Gen X parents. The use cases they highlight—filling out permission slips, shopping for strollers—are aimed squarely at managing the chaos of family life. This isn't another work-focused productivity tool; it's a consumer-oriented agent, and honestly, that’s why I love it.

In this review, I’m going to walk you through my experience setting up Muse, what I had it do, and where it both delighted and disappointed me. We'll cover three key workflows: creating a beautifully designed family morning newsletter, setting and tracking personal goals in a surprisingly gentle way, and testing its ability to shop and buy things online. Spoiler alert: I think this might be my favorite designed agent I’ve played with in a long time.

Workflow 1: The One-Shot Family Newsletter

One of my favorite AI use cases is the morning brief my husband and I started making with OpenClaw. It’s a printout with our schedule, news, and notes for the day that we put on the kitchen table. It’s a great way to stay organized and connect. But getting the design right with tools like Claude or Codex has always been a struggle. The results were functional but... boring. I decided to see if Muse could do better.

Step 1: Giving Muse the Prompt

After connecting my Google Calendar and email (which was a surprisingly smooth and secure-feeling process), I gave Muse a detailed prompt. I didn't hold back.

"What I really want is a PDF agenda for tomorrow morning, kind of like a newsletter, one-page style that I can print out and put on the kitchen table. I want it to have not only our schedule for a week, but highlights of each kid's call-outs for the parents. You might need some information from my email about what's going on, and then recent news, recent events, things in San Francisco, the weather. Um, we can make this kind of a fun read for the whole family that we can talk about, not just the logistics, but also things that are going around in the world, in our city, in our lives."

Step 2: The Onboarding and Information Gathering

Muse’s permissioning model really stood out here. It didn’t just read my email. It asked, “May I go find out some information about you?” After I said yes, it came back with a summary:

• It found my kids’ and husband’s names.
• It identified the kids' activities like jiu-jitsu, basketball, and piano.
• It correctly inferred that we're civically engaged in San Francisco, work in tech, and enjoy personal projects.

It then asked, “Can I use this information please?” This multi-step permission process felt respectful and built a lot of trust right away.

Step 3: The Incredible Result

Y'all, the PDF it generated was beautiful. It was the first time an agent has produced a design I actually like in a single shot. This alone makes me want to explore the underlying Facebook models more.

The newsletter was perfectly laid out and easy to read. It included:

• A weekly agenda: Clearly showing all our appointments.
• Call-outs for parents: It even flagged where Brazilian jiu-jitsu practice and piano class overlapped—a detail every parent can appreciate.
• News to Know: This section was brilliant. It pulled kid-appropriate local and world news (safer e-bike rules, a new telescope observation, the Russia-Ukraine conflict) and included “Talk at the table” prompts like, “What would you name a brand new star-forming nebula?”
• Local Events: It highlighted that the Autumn Moon Festival was happening in San Francisco that weekend.

This workflow demonstrated how an agent can create something that gets you off your computer and more engaged with your family. It was a huge win.

Workflow 2: Setting Personal Goals

One of the most unique primitives in Muse is “Goals.” It’s not just a to-do list; it’s a dedicated space for tracking personal aspirations across health, relationships, finance, and more. I decided to test it with a particularly tricky and personal goal: sleep training my baby.

Step 1: Creating a Goal

I navigated to the Goals section and started a new one with a very human, slightly desperate prompt.

I wanna start a goal about sleep training my eight-and-a-half month old. Gently, we are room sharing because Our house is small, but mom has to get some sleep, please.

Step 2: A Gentle, Empathetic Conversation

The agent’s response was incredibly well-toned. It wasn’t sycophantic like some models can be. It felt kind and human.

“Sleep deprivation with a baby is its own special kind of hard. We'll get you some rest.”

My heart grew 100 times. This is exactly what you want to hear from a personal agent. It then asked for more details about our current routine, and I explained our nightly struggles. The entire interaction felt more like a compassionate coaching session than a chat with a bot.

Step 3: The Action Plan and Reminders

Based on our conversation, Muse identified the core issues (the “1:00 a.m. party” and sleep associations like hand-holding) and proposed a gentle plan, starting with trimming down his second nap.

Once the goal was set, Muse established a reminder to nudge me about it. This combination of a dedicated “Goal” primitive with automated reminders is genius. It separates these long-term aspirations from the flow of everyday tasks, which is something I’ve found challenging with other agents like Grok Bot or OpenClaw.

Workflow 3: Browser-Based Shopping (A Mixed Bag)

An agent’s real test is its ability to take action in the world. For Muse, this often means using its built-in browser. I decided to test its shopping capabilities with two different tasks.

Test 1: Shopping for Shoes (Failure)

My first attempt was asking it to shop for New Balance 9060s. I’m a basic millennial mom, and I need my basic millennial mom shoes. The experience was not great. Muse struggled to find the right colorway and seemed to get stuck during browser use. My go-to agent for this, Codex, would have crushed this task. This was the first major area where Muse fell short.

Test 2: Buying IMAX Tickets (Success!)

Not one to give up, I gave it another, more specific task. I wanted to see if it could find movie tickets. First, though, I had to give my agent, Polly, a makeover. I used a prompt to change her avatar and name:

change your avatar to a teal adorable dragon. I want your name to be Slime, because that's the name of my kid's favorite stuffy

After Slime the adorable dragon was created (complete with custom animations!), I gave him his task:

“Awesome Slime, can you look up when 'The Odyssey' is playing on IMAX in Daly City this week? Preferably during the day when I have a nanny.”

This time, Muse performed much better. Here’s the step-by-step:

1. Search: Slime got to work, complete with a cute animation of him typing on a little fuzzy laptop.
2. Clarification: It came back with showtimes and asked me to confirm which one I wanted.
3. Action: I selected a time and the number of tickets. Muse then opened its browser view, giving me a window into its process without exposing the entire complex virtual machine. It navigated to the theater website, selected the tickets, and proceeded to guest checkout.
4. Payment: It arrived at the payment screen and prompted me to use Stripe Link to complete the purchase by issuing a virtual card.

While I didn't actually buy the tickets (a three-hour movie is a luxury I don't have right now), the workflow was smooth, successful, and a massive improvement over the shoe-shopping experience. It showed that with a specific enough request, Muse's browser capabilities can be quite effective.

Final Thoughts: Design is the Differentiator

After a weekend with Muse, my takeaway is that its greatest strength is its design. The team at Meta has clearly poured immense thought into the user experience. Features like the customizable animated avatar, the trust-building permission requests, and the clean, intuitive primitives like Goals, Ideas, and Library make it a delight to use.

I was especially impressed by the Activity Feed, which provides a transparent, step-by-step lineage of every task, including tool calls and scripts. It’s a feature I wish every agent had, presented in a beautifully accessible UI.

Muse isn't the most powerful or intelligent agent I've ever used. Its browser use can be hit-or-miss. But it is, by far, the most consumer-friendly and thoughtfully designed personal agent I've tested. It never annoyed me, and it consistently surprised me with moments of delight.

I'll be keeping it on my phone to manage wellness tracking, kid activities, and definitely to generate my morning newsletter. For product managers and designers building in the AI space, I highly recommend spending time with Muse. It's a masterclass in how to design agents for real people.

Thanks to Our Sponsors

• Optimizely—Your AI agent orchestration platform for marketing and digital teams
• OpenArt—An all-in-one AI creation platform for images, videos, music, audio, and more