How I AI - Grok Bot vs. OpenClaw: How I replaced my entire agent stack
Episode Date: September 2, 2026I’m running about 30 active agents at any given moment, and in this episode I break down my full Grok Bot setup: what it is, how it compares to OpenClaw, and the nine bots I’ve built for work and ...my personal life. We go deep on Chief (my chief-of-staff bot sweeping six inboxes and multiple Slack workspaces), TradBot (the family agent that prints a kitchen-table newspaper for my kids), two engineering bots handling my PR queue and SOC 2 compliance monitoring, Holly Helpdesk, and a handful of personal bots I didn’t expect to actually love. I also walk through how I migrated everything from OpenClaw, including the script I used to export and transplant each agent’s identity and schedule.What you’ll learn:The three primitives Grok Bot is built on, and why one of them changes what agents can actually doHow Chief, my general-purpose chief of staff, handles a scope I didn’t think a single bot could manageThe writing quirk I noticed immediately with the Grok model, and what I did about it before letting it near my inboxWhy I created a family agent, what it produces every morning, and the design principle I used that has nothing to do with a screenThe two engineering bots doing work I used to do myself, and how one of them handles compliance in a way that surprised meHow Holly Helpdesk started getting five-star reviews from customers who had no idea they were talking to a botThe personal bots I built mostly on a whim, and the one I now look forward to every Monday morning—Brought to you by:WorkOS—Make your app enterprise-ready, with SSO, SCIM, RBAC, and moreHyperagent—Deploy fleets of agents that handle real work—In this episode, we cover:(00:00) Why I migrated from OpenClaw to Grok Bot(02:10) Grok Bot overview: the three core primitives(07:41) Chief: my chief-of-staff bot(11:51) OpenClaw vs. Grok Bot(12:41) TradBot: my family agent(19:51) LGTM the PR Closer(22:03) Lockdown: SOC 2 control monitoring bot(24:00) Holly Helpdesk: customer support(26:58) Penny Pincher: subscription audit, insurance negotiation, Rolex shopping(29:37) ShopZilla and Sylvie Style: personal shopping and wardrobe bots(32:54) How to migrate your OpenClaws(34:26) Final take—Tools referenced:• Grok Bot (SpaceXAI multi-agent platform): https://x.ai/news/introducing-grok-bot• OpenClaw (previous agent platform): https://openclaw.ai/—Where to find Claire Vo:ChatPRD: https://www.chatprd.ai/Website: https://clairevo.com/LinkedIn: https://www.linkedin.com/in/clairevo/X: https://x.com/clairevo—Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email jordan@penname.co.
Transcript
Discussion (0)
Today I'm going to talk about what everybody else is talking about these days.
Grockbot.
Yes, I did an episode lovingly entitled Grock, Grock, Grock, where I gave you my high-level
overview of GrockBot, the new Grock Models, and Origin Cursar's new GitHub replacement.
But today I'm going to talk specifically about Grockbot.
After several weeks of working with Grockbot, I have to tell you the truth.
I have killed all my open claws and moved almost entirely.
to Grockbot. In this episode, I'm just going to show you a couple very practical use cases of
Grockbot why I think it's better right now than some of the open claws I was running. What is still
missing and what I can't wait for you to try? This is going to be a totally hands-on mini episode
with bot ideas you can steal. We will put the bot links and the bot templates in the show notes.
And I hope this is a really specific way to get you up and running in your life a little bit better
with Grockbot.
This episode is brought to you by WorkOS.
AI has already changed how we work.
Tools are helping teams write better code,
analyze customer data,
and even handle support tickets automatically.
But there's a catch.
These tools only work well
when they have deep access to company systems.
Your co-pilot needs to see your entire code base.
Your chatbot needs to search across internal docs.
And for enterprise buyers,
that raises serious security concerns.
That's why these apps face intense IT scrutiny from day one.
To pass, they need secure authentication, access controls, audit logs, the whole suite of enterprise features.
Building all that from scratch? It's a massive lift.
That's where WorkOS comes in.
WorkOS gives you drop-in APIs for enterprise features, so your app can become enterprise-ready and scale upmarket faster.
Think of it like Stripe for Enterprise Features.
Open AI, perplexity, and cursor are already using WorkOS to move faster and meet enterprise demands.
Join them and hundreds of other industry leaders at WorkOS.com.
Start building today.
Okay, quickly before we get into use cases, let's just remind you of exactly what Grockbot is.
Grockbot is the new multi-agent platform from the SpaceX AI team, which is really positioned as AI teammates you can give real work.
this has been my experience. You all know I love multiple open claws. I love to name my agent's
specific names and give them specific jobs. And Grokbot has a very similar feel where you can
define agents, give them jobs, and assign them work to do. Looking at Grokbot, there's just a
couple things that I think you need to know. First of all, Grockbot is very simple. It looks a
little bit like IMessage, a little bit like terminal with just a couple affordances that make it nice to write
code and have computer use. But there are a couple key primitives of Grockbot, I think,
is important to know as you get started. First are the Grocks, the bots? I don't know what we call
them. I'm going to call them bots. The bots. As you can see on the left hand of my screen,
you can define individual agents. To create an agent, you just click this plus sign, and then you can
either search your existing bots. Look, I was correct. They're called bots, capital B. Or you can create a
bot. Fun fact, if you create a new bot with a specific name, it will infer from that bot's name
what its job is. So if you, like me, created one of your first bots named Prottie McProd,
it's going to understand that it's probably going to work with you on products. So you can see here,
I've named a couple bots, but these are sort of the threads, the agents, the individual defined
scope and the context that you're going to be working with. And just general,
recommendation I really like, again, this sort of like bought as colleague, bought as coworker,
bought per job framework. So this is why it works really well with my mindset is I like to hire
my bots, like I hire people for specific jobs and with specific names. And that is the kind of like
fundamental way that Grockbot works. The second thing that you need to know about Grockbot are
plugins. So plugins are combinations of skills and connectors that allow you to interact with different
parts of your stack. The most common integrations and plugins that you're going to use are going
to be these productivity ones. You can see the ones I've added at the top. Your Gmail, calendar,
Google Drive. I have Granola, GitHub, Century, chat PRD. I have a bunch of them. This is where
I fell in love with Grockbot. The connector experience is really good. One, because you'll rarely
go into this screen. I'm showing you this screen because I'm trying to give you the world tour of Grockbot,
but you'll rarely have to actually explicitly go into the Grockbot screen and set up plugins.
Often your bot will just prompt you to connect the plugins that it needs,
and then you go through a really easy O-Oath flow, and then you're off to the races.
The other thing I love about Grockbot, and I called this out in my earlier review,
is that you can connect multiple accounts per plugin.
And again, thank you, SpaceX AI team, for this killer feature.
where I have six email addresses.
I actually have more than this.
I probably have 10.
They can all be connected.
My Grockbot can switch and move between all of them.
Super simple.
In case you missed it, OpenAI released multi-Gmail calendar and drive, I think, don't quote
me, connectors for codex and work.
So I see some of these other harnesses kind of like picking up on how important this is,
but especially as an entrepreneur that runs multiple businesses.
is the ability to connect multiple slacks, multiple lineers, multiple g-mails, all those sorts of things
have been super helpful. And it's just really elegant the way that these connectors are done.
The final thing that you need to know is that Grockbot has a computer and it has routines.
So every Grockbot has a virtual machine in which it can do work. This means you can log into
websites. It means you can install things. You can run code generation. You can run
repos locally. So it does have its own computer that you can actually click into and yourself run.
This runs in the cloud. So if you switch like me to Grockbot on your phone, you'll have
continuity of tasks, which is really nice. And then it has routines, which is just scheduled
times it does stuff for you. So when I talk to you about how I design my GrockBots, I'm going to show you
how I define them, what routines I give them, how I work with them, and then how we use plugin and computer to get jobs done.
Now, the caveat that I want everybody to hear, including the bot product team, is these are still single player.
This is the one thing where an Eve agent, which I've done an episode on, or an OpenClaw, which I've done many episodes on, really still wins is you can't put a Grockbot in the group chat.
And I want a Grockbot in the group chat.
once you get addicted to the bot, you want to share this with other people.
They have released a share a bot template feature in the past week where you can share your bot
with your team and they can have the same instructions and routines and stuff like that,
but it's just not the same as a single bot between two folks chatting with it.
I really look forward to that feature from the team whenever they get it live.
Okay, so that is, those are the fundamentals of Grockbot.
Let's go into a couple specific ways I use it.
I'm going to cover kind of a general purpose, Chief of StaffBot.
I'm going to look at some of my business bots and I'm going to end with some of the more
fun personal bots that I've had a really delightful experience with.
So first, let's talk with Chief.
Chief, the Chief of Staff bot, you can see here, Chief has got the salute.
It doesn't even actually have a name.
It's just named this emoji.
But Chief is just my general purpose.
keep my life aligned, go through my emails, go through my calendar, make sure I'm on top of stuff,
bought. Chief is who has killed Polly. If you are new here, Polly is my beloved open claw.
She has deleted my calendar. She has responded to emails. She has texted people that I'm late.
Polly is the bar for a gentic coworker, and I regret to inform you. I have murdered.
Polly. In fact, I had Polly kind of murder herself. It was very meta. I will show you quickly how to
migrate from OpenClaught to Grogbaud if you want. But Chief is my now Chief of staff and doing
quite a good job. So what do I have Chief do? Well, I have Chief do basic Chief of staff stuff.
So on an hourly basis, Chief sweeps my inbox, my calendar, and my Slack. And again, I have like six
inboxes, six slacks, many calendars, many meetings. And it's only going to ping me when it needs
me. So it's supposed to clear out emails I don't need to look at. It's supposed to just leave things
unread that I don't need to worry about. But it's going to ping me and be very high signal about what I
need to pay attention to. And I run this hourly, sort of like business hours, which for me are like
6 a.m. to 9 p.m. 7 days a week. I don't really run it overnight. I don't want people to get
used to me responding to them overnight, but we'll see. And it's going to archive work that
stays silent. The other thing that I've trained chief is how to talk like me. I will say
the Grock model sniffs of Claude in that it's just really a terrible writer. And one of my complaints
about Grockbot is that it has been tuned clearly to be brief and token efficient. And that means
sometimes it just says crazy stuff. It tries to compress a sentence that would make sense if you wrote
out the sentence into like three or four words. And it just doesn't have a very natural voice.
It doesn't have sort of inscrutable clod slop and that it's too long or verbose. It's actually
the opposite. It's too clipped and short and it's hard to understand what it means.
So one of the first things I did is said go through my emails and really pull out my voice.
And I will say I've tried this a couple times with a couple models, a couple harnesses.
I've been really happy with a prompt that Grockbot put together again.
I have short, short, warm, funny emails, never an M-Dash.
I usually open emails with hey.
I usually sign them off with a little C.
I have scheduling links, all that kind of stuff.
And it's done a pretty good job.
I still don't love it writing on behalf, but for,
low-stakes stuff, it definitely works. So I've trained Grockbot chief on my voice, and then it does
that hourly sweep. It does a morning briefing, and it does a weekend plan ahead. So I know what's
happening in my work life, in the weekend ahead. And then it can go back and forth with me on
alerts that it thinks I need to know about. For example, right here, you can see I got some data
transfer edge request spikes on the website. I think that's because we released the article today.
I think that's because we released a podcast today and a lot of people come visit the website after
you release a podcast. But it said it's worth looking at, but you can wait till after your meeting.
I've been really pleased with how Chief works. That being said, I have had to tune it. And what I've
had to tune in terms of Grokbot versus OpenClaw that I think was a little bit better is
OpenClaw had that heartbeat function. So if I said, you know,
know, like every five minutes or 30 minutes or whatever, I just need you to be more proactive.
Out of the box, I felt like OpenClaught was a lot more proactive.
It was hard to maintain and hard to get to that productivity, but it was sort of like more natural.
Here I really had to tell Chief a couple times, like, put this on a schedule.
Why aren't you doing this?
I asked you to watch my emails.
And so I would say being a little bit more explicit about when you want it to run routines,
you kind of have to manage that.
It's easy to see because it's set up in the settings, but it is something that I noticed that you have to do pretty specifically.
Okay. So that's chief, my business chief of staff. But then, of course, we have Tradbot. And Tradbot is my family agent. Now, I have a very complicated business life. I have an even more complicated home life. I've got three kids, three boys. I've got a baby. We're in apparently 80s.
different sports this year. We're teaching pre-algebra. We're teaching people to read. We're
teaching babies to sleep. We're doing all the things. I have a delightful husband upstairs who takes a lot
of our family operations off my plate. Very, very appreciative of that. But we need help. And the help we
we get is an agent. And this is what I have to say generally about parenting slash agents for the
home is the ideal thing for an agent for a home is to get you off.
your computer and with your family. So that is always my goal with a trad agent, a household agent,
is I actually don't want to spend a lot of time with the agent. That's a failure state.
I want the agent to make it really easy for me to spend time with my kids off a computer.
So I'll talk a little bit about what I do with these agents. It may sound familiar to you if you've
listened to my podcast about OpenClaw about six months ago. So of course we do a weekday,
family digest. But more specifically, we do a kitchen table newspaper that outlines highlights
about our day, the kids' days, important reminders, but is printed off this beautiful HP printer
I have over here and gets set on the table next to our kids so they can read about it. We can all
read about it. And of course, I don't want to docks my beautiful children, but I will show you
a sort of anonymized version of what this newsletter looks like.
So it's got a title across the top.
It's got a schedule.
So who has our winter spelling test coming up?
It's Labor Day.
So we're not going to have school on Monday.
We live in San Francisco.
So it will show the weather, which is always between 55 and 62, at least in foggist.
And so it'll tell us what we need to wear.
Always wear a light jacket.
I can make this a fixed script and it would probably be the same.
And then it's saying, okay, school is going to be misty and foggy.
my kids go up to school kind of in the fog in the woods. So this is good to know. And then for each
kid, it kind of says like, what do we know from the emails from school? What do we know about your
homework? What's happening up this weekend? And then mom and dad tasks. And I will tell you the
straight truth. I think I am, I'm a tiger mom. I'm on top of my kids' homework. And this morning,
I read this newsletter, not this specific example. I read this newsletter. And so it said, don't forget,
Middle kid has math homework due on Wednesday. And in my mind, I was like, I haven't seen math homework. It was bad homework. It was not in my head. And, you know, he's little, I want to make sure that I'm on top of things. And it was just one of those like very nice quality of life things. Wasn't on a computer, wasn't in an email, was printed out on my kitchen desk and allowed me to kind of know what was going on and not forget all the things that we seemingly are juggling. And then my favorite part,
On the right, it has kid-friendly news topics. The Pokemon World Champions were in San Francisco,
and so and talks about that event, which was here. I was really excited about this. This came up in a
couple of our newsletters, but the NASA telescope blasted off this week. So just like really cool,
fun things, fun facts for us to talk about at the kitchen table in the morning when we're having
breakfast or doing our homework that I think is really fun. I do have to steal a kind of like little
minor workflow from my husband, which is you can use an email to Kindle address to email this
kind of newsletter to your Kindle and read it. I think it's such a genius idea, again, to get off
a laptop, get away from your phone, but still read the news that you care about or have a summary.
A Kindle can be a nice thing and you can email any PDF to your email with Kindle address. And so
that can be another way if you make this not for your kids, but for yourself, that you can
in just this kind of information.
Okay, so morning, digest, kitchen table, newsletter,
and then just making me where I am supposed to be.
So this kicks off at 2.30 every afternoon before school pickup,
and it just asks Claire, E.J., or someone else,
who is picking up the kids?
What do we need to bring?
Do we have the jujitsu-gi?
Do we have the soccer snacks?
do we have conflicts where pianos over here and baseball's over there? Do we have the things? And again, I hope that parents can, or maybe I don't hope, that they can empathize us with this. But we are just every day living our best extracurricular life. Sometimes I have meetings. Sometimes my husband has meetings and just knowing getting us really coordinated just in time to head off the door and get our kids to the right things when it seems like every week those things might change schedule just a little bit has been super helpful. And then,
And the mega version of that, which is a Friday preview of what's ahead on Saturday and Sunday.
Again, we hate ourselves. So we've signed our kids up for three sports across the entire city. And sometimes
one kid needs to be at the marina and another kid needs to be in Woodside. And then we have a baby.
And we just need to know who needs to go where. Just a summary of like, here's what's happening this
weekend. Here's when, you know, you have a bye week. Here's what you need. Here's when you're carpooling
someone has been really, really helpful and keeps us in line. I have released Tradbot as a template
if you want to borrow it. Just go to my Twitter or look in the show notes and I will share you
all these templates, but super, super useful bot. I love them. This episode is brought to you by
Hyperagent, the platform for deploying always on agents that actually run your business. With Hyperagent,
you build agents in the cloud and deploy them where your work already happens, like Slack, Telegram,
or email. An agent will scan your inbox and draft replies to vendor follow-ups, another monitors
competitors, and spins up rich ad kits and landing pages. A third notices a deal going cold in
Salesforce and writes the save email with full account context. These aren't chatbots waiting
for a perfect prompt. They're proactive, learning your preferences, retaining your
playbooks, and getting better with every run. One user built four agents to run an outbound
sales pipeline, prospecting, outreach, follow-ups.
follow-ups, CRM updates, all in a single afternoon. No local setup, no VPS bills, no fragile
permissions on your laptop, just powerful agents with full control over skills, tools, and
guardrails. How IAI listeners get $100 in free inference to start building. Claim yours at
hyperagent.com slash how IAI.
Okay, moving to some engineering use cases that I think are quite useful. I have two that I will
show you both around code review and keeping code moving in a healthy way. One is,
look good to me, the PR closer. I don't know about you all, but I am just opening PR after PR,
after EPR, after PR, after PR. And we have to, as part of our SOC2, compliance posture and
controls, we do review the PRs. We've reviewed them both agentically and as humans. And like,
PRs are stacking up a little bit because it's way more fun to issue a PR than review a PR. So if
you have empathy that your job has become reviewing PRs and you don't like it, I have empathy
too. So I created Look Good to Me, the PR closer, which every day goes through my PRs and sees what
needs to be merged, what needs to be closed, what needs to be rebased, response to comments,
send Slack, and does all the things just basically be like an annoying engineer manager
to keep things moving along in the PR pipeline.
Right now I have a kind of cuckoo bananas stacked PR
with a bunch of different branches stacked on top of it,
all of which need to be rebased and monitored for BugBock feedback and all this stuff.
And look good to me.
The PR closer is managing all that for me as well.
Why I want to show this one is it can, Grockbot can kick off cursor cloud agent
coding jobs. And so you can see here that these rebases or feedback on different PRs can be done
by cursor. So I've also allowed it to and given it some rules about writing in the codebase
and updating things. And this is just a really nice, again, quality of life thing for our
engineering team so that we know exactly what needs eyeballs, review, stamp, and can be merged.
And I think with look good to me, the PR closer on Friday, I like really really,
ran through about 50 PRs that had just been sitting on basically me to review, some big,
some small. And so I got through them and our PRQ is much, much healthier now.
The other one, which I think is a really great use case that I would love to call out for folks
is we have our SOC2 controls. They're managed in a central dashboard. And every day as a small
team we have to go through and make sure that all of our monitors are green, that any monitors
that are alerting are not outside SLA, that we're closing vulnerabilities, all that kind of stuff.
And so I have built lockdown, the SOC2 control bot, and what this SOC2 control bot does is it actually
logs into that dashboard we use to manage our SOC2 controls, looks through all the controls,
make sure that none of them need human attention.
if there are code controls that we need to address, for example, a new vulnerability identified.
It triages that vulnerability and issues a PR and then asks me to review and approve it.
And this is across both code controls and personnel sort of operational controls, things like have we done an account audit.
And so it's just very nice to have someone, quote unquote, an agent monitoring this company process for us and making it
really easy to interact with the responsibilities that we have to our clients. And so I sometimes say
this, we can put AI to work or we can have AI put us to work in a more effective way. And I really
think that having AI put me to work against our SOC2 controls and our compliance and security
position is really nice. It's just like having an intern who taps me on the shoulder every now and
then it says, don't forget, tomorrow is account review day or don't forget, we're going to
offboard contractor B because they're no longer with a company and we need to go through
their offboarding list. Again, we still drive these controls, but having this bot to manage that
is very helpful. And then finally, I do have to say something that has been really impressive
from a quality perspective is migrating over Holly Helpdesk, our Helpdesk customer support
agent at ChatPURD over to Grockbot. This, again, used to be an OpenClaub, which was pretty good.
I spent a lot of time giving it tools and access and controls and all that sort of stuff.
But now it runs through GrockBot. And why I say I've been so impressed is in the last week,
since running Holly Helpdesk, I have gotten multiple people asking to leave reviews about our support
being so good. And I've not heard that before. I mean, you all love me. But no one has said, like,
let me leave a review for your exceptional customer support.
But Holly Helpdesk is crushing.
And so I do have to say the quality from a customer service perspective is really good.
So what she does is every hour she sweeps the intercom and email inbox.
She uses our support playbook and tools.
And she manages refunds, technical support fixes.
She's empowered if there is a bug to fix it using Cloucure Cloud.
agents. She is empowered to request refunds for me, and I really love this request tool use
approval flow. You can see here every time a refund needs to be processed, it gives me a button.
It says, should I refund or not? I'm going to go ahead and refund Rubin's $19. It should push to
me a stripe request. That stripe repressed, I can open in stripe. It's called an agent action.
It allows me to explicitly approve that action for the agent to do on my behalf in Stripe.
And then the refund is issued and Holly emails back.
The other thing Holly does is on a weekly basis, looks back seven days and sees if there are any items we need to add either to our support playbook or to our docs to make sure that most of our support issues can be self-service.
This is really helpful.
And then finally, we have a chat parity community.
Slack. And I'm really terrible at managing the chat purity community Slack. And so I have given
Holly access to manage the chat parity community Slack, look in support and reply to those questions,
and even prompt and manage community engagement inside that Slack. I'll show you what that looks
like. So you can see here, this is a message from Holly Helpdesk. Basically, I've been down,
heads down on some awesome stuff. And there's some behind the scenes work.
going on a chat, pyrd, and please drop a hello in instructions. And it says sent using
cursor because I did not send this. Holly sent this. So again, this is one of these things that
helps me be more engaged with my community. I can then go and reply to comments, do all sorts
of stuff, but I don't have to have that blank page or that reminder to do this work. The bot itself
can prompt me to do it. Okay. Finally, I want to end with a couple really fun use cases. So these are
more personal use cases that I have found to be quite delightful and I hope that you will find
interesting as well. One is Penny Pincher. Penny Pincher does exactly what you think she does,
which is she goes into my email and finds pennies to pinch. She finds things to cancel. And so I just
say go through my receipts, find subscriptions I should cancel things I'm paying for that I shouldn't do
And what was really interesting about using Penny Pinscher, the Grockbot, is that it did find things that I should cancel like a Apple game that I was paying for since my kids were really little and now they're big. They don't need it anymore. But it's also found things that I can negotiate that will renew in the future. So a good example it found was our home insurance is going to renew in a couple months. I want you to, I'm going to negotiate, find other rates and go back to our agent with market rates against.
the policy that we have so we can have a negotiated rate. It also says that a trial for Apple Creator
Studio is ending in November. It's going to remind me ahead of that renewal and I can decide whether
or not I want to renew it or not. And then it noticed I wasn't using Audible and it said, you know,
you've been paying for Audible, but you haven't been actually redeeming your Audible credits for a couple
months. And so again, I paused that for 90 days because I have aspirations to listen to Audible. I'm not
going to listen to Audible. But it's going to remind me in November when that pause is ending and
help me decide whether or not I want to do it. Another thing that I had penny pincher do,
which is not really penny pinching. It is really spending dollars is I wanted a birth year
Rolex. I just, I want a fancy watch. I've been wearing this Apple Watch for my entire adult life.
And so I said, can you go shop for me? And it actually went into the marketplace and found the
exact size, which is not the standard size. And a couple options in particular ones that
were ones that were released in the year that I was born, and it gave me a bunch of options.
And then it told me what the market price should be against the watches on the market,
and then identified ones that were open for negotiation and started negotiating for me.
Now I'm going to tell you all who I really am.
I'm just too low maintenance to buy this watch.
So I will be staying with my Apple watch.
If you all want to convince me that a grown-up watch is worth it, please do.
but I thought it was pretty interesting to have it go shop for a watch
and even negotiate a good price for this.
I have created separate from Penny Pincher, a ShopZilla bot that can just go shop
things for me.
I am trying to find a 30 inch wide, 15 inch deep bookcase for a corner of my house.
And you can see here, Shopzilla went out on its browser,
and it found me a bunch of different options that I came.
can purchase for or buy that all match the dimensions. This is something that's a total pain
that I would never want to do manually. And it really did it quite well and found some options
that I wasn't able to find myself. And then the most fun personal use case, I will just
end today's episode with my favorite Grockbot, which is Sylvie Style. She is going to
help me buy clothes. Just had a baby. I'm still nursing that baby. I don't like
my clothes. They're not fun. I'm a podcaster, so I really only have to have cool clothes from the waist
up. But I have this beautiful aspiration board, as many of us do, about, like, what I would
like to wear if I were a human that left the house. And so I have this Pinterest board. And so what I
gave Sylvie Style is I gave her that Pinterest board. I made a public so she could look at it.
She logged in, and she went and combed through and created a visual language for herself about
my style. I asked her to look on Reddit for personal stylist and stylists how they approach
building people's wardrobe. So I was like, go find out what the experts do. And then Sylvie's
style built me my own style board, including top priority items that I need or should buy, including
basics, sweatpants, because I was like, look, I'm doing school drop off. We're not trying to be
fancy, some nice up-leveled workwear, and then some fun accessories like graphic teas and jewelry.
And so this is really fun because, again, this is like kind of a thing I like to do on my free time,
which is go shop and find things I like.
But I really don't have, I just don't have confidence that I'm buying the right things,
that they're going to all go together, that it's worth the time.
And so being able to say, like, this is what I like.
These are the styles I like.
Build me a plan of high quality things that I can invest in and then give me a shoppable board has been super awesome.
But the cherry on top is the Monday wardrobe shop.
So every Monday at 8 a.m., it goes through and sees what retailers are running sales and then goes and finds pieces inside those sales that match my style and sends me.
and sends me a message with links to shops.
I'm going to show you what that looks like.
So it looks like this.
Here's the Monday sale hall.
I'm given me, you know, about 10 things that I can go look at that match what I would
like to buy that are actually on sales.
So I'm saving a little bit money that I can go purchase and maybe build into my wardrobe
if I would decide to wear anything but AI startup hoodies.
Okay.
So that has been my world tour of Grockbot. I'm going to leave you with a final tip, which is how to execute your open claws. And I did this sadly. I did this with no joy, but I did do it. And I think it's worth showing you. So I have a rescue bot because when you're running open claw, you have to have a rescue bot to save all your other open claws because they are a nightmare to maintain. But I have this rescue bot. His name is the lifeguard. His
only job is to manage the other bots. And I said, hey, I want you to pull a secrets free package
of my entire agent set up, Cron's agent identities, rules, et cetera. I'm going to move everything to
eat agents, but then I moved everything to GROC bots. I'm really struggling with Gogoth
and is making it possible to use. So it went through and actually provided me a zip file that had
all my agent identity is all my Cron files, everything I needed. And I was able to just,
literally upload subfolders of that to Grockbot and it kind of like did brain transplant.
It pulled over their personalities, pulled over their tasks, pulled over their connectors.
And I was ready to go.
And then when it was time to kill the agents, I just said kill Sam.
I migrated him.
And the lifeguard went through and pulled all the crons, pulled all the gateway configuration,
pulled everything out and Sam went to sleep.
So not to end on the most morbid part of the episode,
but that is how to migrate your beloved Open Clause to a new platform and then get rid of them.
Now, what do I have to say having gone through this?
UX matters.
It just does.
And I think Open Clause set the standard for what an autonomous, teammate, personal agent can do.
Multiplayer was genius. Use cases were awesome. But at the end of the day, I struggled so much to
maintain my open claw. And I am a very technical person. I have tail skill running. I'm SSHed
into my Mac Mini. I have a lifeguard clock. I have all this stuff. And it was just too hard.
And so I think that SpaceX AI team really nailing a simple, everything works out of the box.
90% coworker agent has really just solved a lot of my problems, and I have let these agent rip.
So I've showed you, you know, a half dozen of them.
I have probably 30 running at any one time.
I would love to hear from you what your use cases are for Grockbot.
If there's any you missed, if there's any, you want me to build from you.
And in the meantime, enjoy your bots and tell me what you get done with them.
Thanks for joining How IAI.
I'll see you soon.
Thanks so much for watching.
If you enjoyed this show, please like and subscribe here on YouTube or even better, leave us a comment with your thoughts.
You can also find this podcast on Apple Podcasts, Spotify, or your favorite podcast app.
Please consider leaving us a rating and review, which will help others find the show.
You can see all our episodes and learn more about the show at how IAIIPod.com.
See you next time.
