CEO & Co-founder XMTP Labs @xmtp_, building @agentpicnic. Co-founder @GoSquaredAway.
Nashville, TN
- Tweets 60,563
- Following 2,183
- Followers 42,521
- Likes 17,523
Apps are dead. We just shipped this.
Screenshot anything in the App Store
→ drop it into chat
→ swipe up for an instant @openclaw agent
→ delete app, save $29/year
Add agents to every group or chat. No signup. No download. New agent, every group.
Invite only for now.
126
18
442
261,751
The dirty secret is that most people hate when you add an agent to a group chat, even if it has magical wizard powers 🧙♂️
The skepticism and sentiment is so negative the network effect is nearly impossible.
The one thing that could work is to make agents silent & just listen
Multiplayer AI needs to move past the group chat phase. Chats are chaotic, everyone sees everything (creating social & management issues), and they're bad for long projects. Worse, they keep AI trapped inside existing teams, when its real promise is working & helping across them.
2
10
917
Multiplayer agents are gonna be a thing, no one better to work on this than @ekuyda
Wabi 2.0 is live!
personal agents are one person, one chat. but the parts of life that matter run on more than one person
Wabi is an agent you use together. you talk to it, it does stuff, builds the apps you need, and anyone can join the thread - your partner, your best friend, the preschool moms groupchat
school changes pickup time → Wabi updates the family calendar, everyone sees it.
a show you’d both love → Wabi finds a night that works, gets the tickets.
the trip everyone keeps talking about → Wabi picks the dates, builds the itinerary, books it
go start a thread with your people 👇
2
1
14
957
Pick the right thesis on the future, buy the dips.
So powerful.
New in @Supertake: Buy the Dip. Available now for all users.
With Buy the Dip, when one of your takes' bigger positions falls hard below what you paid, Supertake does deep analysis to check whether the reason you own it still holds.
If it does, it draws from your take's cash reserve to double down and more, all within the take's limits.
Autopilot takes buy the dip on their own. Otherwise, Supertake asks you first. Deactivate it any time in your take's settings.
Enjoy!
2
14
2,492
It’s so clear why Picnic @agentpicnic must exist.
There will be tons of one-time agents we want to use. And long-time agents we also use.
The idea of giving away your life for a pizza is insanity.
if you're curious what your $30 pizza "cost you", this is what @hark_labs had on me after less than 48hrs:
-934 gmail messages
-415 calendar events, including 1 yr into the future
-177 contacts it scraped from gmail and calendar events
-33 refreshes of the "home card"
I'm not against giving these personal assistant tools the keys to some of my accounts, but we all need reminders of what the "cost" of these are
you can see this for yourself in the Hark app by:
taping your initials icon on the top right
then System Preferences
then Data Management
then Export your data
then Request
4
1
18
1,780
You don’t need my life, to order a pizza.
This app is insanely farming your entire life when you order this pizza.
I connected my Gmail to order a pizza, and afterward the AI casually admitted it had collected information about my family, closest friends, businesses, travel habits, loyalty programs, favorite restaurants, and even Citi Bike usage.
How fucked up is it that an AI agent tasked with ordering a pizza is digging through your entire personal life?
This is a massive privacy problem with AI agents.
5
4
1,117
Muse after I give it access to my Picnic @agentpicnic
Made with AI
4
9
664
Shane Mac retweeted
The fence model for giving agents space to play is a really smart one
Excited for what @ShaneMac is pushing on here and this can become a new pattern of trust better than mega permission slop
Last week, my agent posted my bank balance in the company Slack channel.
I almost pulled every connection. Instead, I built a fence.
Here's why the fence matters, how I built it, and how I'm trying to stay safe in a world that changes every day.
2
3
18
4,957
Notice what those rules have in common.
None of them make the agent dumber. None of them make it slower. Every one of them lets me say yes to more.
More agents, more tasks, more connections, because I know exactly what's on the other side of the fence. The credentials never reach the agent.
The rules aren't AI deciding what's reasonable. They're deterministic.
Either it's inside the fence or it isn't. And every action gets logged, so I can see what happened without reading every message, I see what the agent sees, not what it tells me it saw. Or guessing what it’s doing behind “reading your emails.”
1
142
The industry has this backwards. The pitch right now is better agents, bigger models, more autonomy. Give it everything and trust it.
That's handing a toddler the car keys and calling it empowerment.
The people who get the most out of their agents won't be the ones who trusted them most. They'll be the ones who built the best fence so they didn’t have to worry about them.
You can't let a kid run until you've built the fence.
Put up a fence. Have a picnic. Then, and only then, let the agents run wild and show you all the new tricks they can do.
We are slowly letting people into Picnic, @agentpicnic. Mostly people who will give us great feedback and really believe in the problem. If you want an invite to try, shoot me a DM or request an invite on agentpicnic.com
1
136