CEO @builderio - visually code together
San Francisco, CA
Joined October 2008
- Tweets8.4K
- Following804
- Followers139K
- Likes21.5K
Pinned Tweet
Introducing Agent-Native: an open-source framework for building agentic apps.
Agents love walls of text. But humans don't.
An agentic app let's you interact with agents through a visual interface. Engage with agents visually, via text, or both at once.
Now in beta.
Steve (Builder.io) retweeted
The first Builder × Agent-Native hackathon is a wrap.
Five winners, $10,000, and just 0.2 points between first and fifth.
Every winning app had one thing in common. The AI takes the grunt work, and the important calls stay with you...
Steve (Builder.io) retweeted
Eighty came 2nd at the @builderio Agent-Native Hackathon with $2500 prize and also got recognized as Best Use of Agent-Native 🥈
Eighty helps people on Medicaid prove their 80 hours of work a month, straight from their phone.
You send a voice note or a pay stub photo on Telegram. The agent reads it and adds it to your month. Tap "Do my month" and it checks everything against the rule, asks the one thing only you know, and builds a proof packet. Nothing counts until you confirm it.
Agent-Native is what made this work. The web app, the Telegram bot and Claude (through MCP) all run on the same actions. So the agent and the person always see the same month, and a voice note shows up on the web page while you're still in the chat.
Thank you @builderio for the hackathon and the framework🙏
tryeighty.space
In 2018, Arkansas cut about 18,000 people from Medicaid. Most of them qualified. They lost their coverage to paperwork.
On December 31, 2026, a rule like that goes national: 80 hours of work, school or volunteering, or $580 of pay, every month, with proof.
So I built Eighty. It does that paperwork from the phone people already use.
Send a pay stub photo or a voice note in English or Spanish on Telegram. Eighty reads it and proposes a line. Your month fills in on the web as two rings. Tap "Do my month" and it asks you the one question only you can answer, then gives you a proof packet, cited to the law, ready to send to your state.
You can even connect Claude and ask what's left in your month.
The agents propose. Only you confirm. Nothing counts until you do.
Why it only works on Agent-Native: Eighty has four ways in. Your buttons on the web, the helper on the web, Telegram, and Claude. All four run the same 27 actions, on one database, with the same check of who is asking. Built any other way, that's four separate systems that could disagree about whether someone keeps their health coverage.
And the rule that matters most is set on the action itself: the agents propose, only you confirm. No agent, in the app, in Telegram or in Claude, is ever given the confirm button.
Built on Agent-Native for the Builder.io Agent-Native Hackathon.
Try it, no sign-in needed: tryeighty.space
Telegram: @TryEightybot
Code: github.com/Jennycruzy/eighty
Introducing Agent-Native Clips, now in beta.
Give any agent eyes and ears in a click.
Record your screen, meetings, or voice, then share a link so agents can see and hear everything.
It’s free and open source.
Clips also:
- Searches every word across your library
- Answers questions across all your recordings
- Records meetings and writes the notes and action items
- Lets you dictate into any app on your Mac
1
2
734
Clips is free and open source. Use the hosted beta, or host it yourself.
It runs on Agent-Native, so your agent can do anything you can do in the app.
clips.agent-native.com
5
735
Steve (Builder.io) retweeted
In 2018, Arkansas cut about 18,000 people from Medicaid. Most of them qualified. They lost their coverage to paperwork.
On December 31, 2026, a rule like that goes national: 80 hours of work, school or volunteering, or $580 of pay, every month, with proof.
So I built Eighty. It does that paperwork from the phone people already use.
Send a pay stub photo or a voice note in English or Spanish on Telegram. Eighty reads it and proposes a line. Your month fills in on the web as two rings. Tap "Do my month" and it asks you the one question only you can answer, then gives you a proof packet, cited to the law, ready to send to your state.
You can even connect Claude and ask what's left in your month.
The agents propose. Only you confirm. Nothing counts until you do.
Why it only works on Agent-Native: Eighty has four ways in. Your buttons on the web, the helper on the web, Telegram, and Claude. All four run the same 27 actions, on one database, with the same check of who is asking. Built any other way, that's four separate systems that could disagree about whether someone keeps their health coverage.
And the rule that matters most is set on the action itself: the agents propose, only you confirm. No agent, in the app, in Telegram or in Claude, is ever given the confirm button.
Built on Agent-Native for the Builder.io Agent-Native Hackathon.
Try it, no sign-in needed: tryeighty.space
Telegram: @TryEightybot
Code: github.com/Jennycruzy/eighty
Steve (Builder.io) retweeted
For the Agent-Native hackathon on @builderio , I built and Submitted Stigi ;The feedback board for product teams
Drop in hundreds of customer comments and an agent sorts them into themes. Drag two notes to correct it, and it re-sorts the rest to match.
Product managers and researchers struggle to turn hundreds of raw customer comments into themes. One-shot AI summaries fail because you can't correct them: fix one grouping by hand and the rest of the pile stays wrong.
Stigi lets you correct the agent's sorting by dragging two notes together. The agent names the theme, pulls in every matching note from across the board, and writes a one-line reason on each note it moved.
How it works
1. Load 200 sample comments (or import a CSV). The sample board starts imperfect: six clusters and 50 notes unsorted.
2. Drag two related notes into a new cluster, for example "which plan am I even on?" and "the billing page makes no sense".
3. The agent names the cluster and moves the matching notes in. In my recorded run it moved 14, each with an "Agent:" reason chip.
4. Click Sort with agent to file the rest in batches of 25, then Report for themes, quotes and a next step you can edit and export as Markdown.
- Corrections that stick: one drag teaches a theme, with no prompt to write.
- A reason on every move: the agent can't move a note without a one-line reason.
- An insight report: themes by size, verbatim quotes and a suggested next step.
- Safe to try: sign-in required, one private board per account, 40 agent turns a day.
How it uses Agent-Native:
the board UI and the agent call the same defineAction() actions (one move-notes for a drag and for an agent move, which must give a reason), share your selection through app-state and view-screen, and the agent opens the report with navigate.
Pre-existing work and what changed:
built on the Agent-Native "chat" template and toolkit (@agent-native/core 0.226.0) plus open libraries; everything Stigi-specific (actions, schema, board UI, agent skills, turn cap, landing page, logo, deployment) was built during the event on 2026-11-09.
App: stigi.suitz.xyz
Code: github.com/Dotman-Bei/stigi
Slide: stigi-slides.vercel.app/
Feedback welcome, especially on whether dragging two notes feels like enough to teach the agent a theme on your own data.
Below is the 1:38 minutes demo walkthrough @Steve8708
Steve (Builder.io) retweeted
𝐏𝐀𝐕𝐈𝐋𝐈𝐎𝐍 is a floor planner for events where event organizers and Ai agents collaborate to design, optimize and manage venue layout in real time.
It allows users to create scaled floor plans, arrange seating , stages and equioment and explore layouts in 2D or 3D. It’s intelligent layout engine automatically checks spatial constraints , seating capacity, safety requirements and costs ensuring every design is pratical and ready for execution.
Rather than making changes without permission, the ai proposes improvements that users can review, compare and approve, keeping humans in control throughout the planning process.
𝐇𝐎𝐖 𝐈𝐓 𝐔𝐒𝐄𝐒 𝐀𝐆𝐄𝐍𝐓-𝐍𝐀𝐓𝐈𝐕𝐄
- One action layer: the canvas and the agent call the same defineAction() contracts (move-objects, add-objects, propose-layout, export-setup-pack and others). There is no agent-only path.
- Application state: navigation and the current selection are synced, and view-screen returns them verified, so "move this closer to the stage" means the chair you selected.
- A human gate in the contract: apply-proposal and restore-version are UI-only actions, hidden from the agent's tools and MCP. The agent proposes; a person applies.
- Every write is versioned and audited in PostgreSQL through the framework's database conventions. Manual editing works with no model key.
- Each visitor gets an isolated guest sandbox with rate limits, so the live demo needs no sign-up.
Pavilion : pavilion-nine.vercel.app
GitHub Repo: github.com/Enoch208/pavilion
𝑯𝒆𝒓𝒆 𝒊𝒔 𝒂 2 𝒎𝒊𝒏𝒖𝒕𝒆 𝒗𝒊𝒅𝒆𝒐 𝒘𝒂𝒍𝒌𝒕𝒉𝒓𝒐𝒖𝒈𝒉 𝒐𝒇 𝒉𝒐𝒘 𝒕𝒉𝒆 𝒑𝒓𝒐𝒅𝒖𝒄𝒕 𝒘𝒐𝒓𝒌𝒔
@Steve8708 @builderio @tempoimmaterial
Steve (Builder.io) retweeted
we built strip: turn a boring changelog into a 4-panel comic, made with an ai agent.
paste the release note. pick one of three angles. edit a line, stamp it mine, and the agent can't touch it.
built on @builderio's agent-native. same action for buttons, in-app agent, and mcp.
live: strip-kappa.vercel.app
code: github.com/emmagh1/strip
Announcing the Agent-Native hackathon!
$10k in prizes. Come build agentic apps with us
builder.io/hackathon
Steve (Builder.io) retweeted
I built Groundswell for the @builderio Agent-Native Hackathon 🌊
It reads your GitHub issues, groups them into the problems users keep hitting, and ranks them with real quotes. Drag a card and the agent learns.
Tested on 340 real issues: 96% sorted.
github.com/Pavilion-devs/gro…
Steve (Builder.io) retweeted
I built Apprentice for the @builderio Agent-Native Hackathon.
Triage issues as usual. It learns from your clicks, backtests itself on your history, turns each correction into a test, and earns autonomy on a trust ladder.
Show it once. Delegate forever.
apprentice-agent.pages.dev
Steve (Builder.io) retweeted
so I built Returnly for the @builderio hackathon
you know how you buy something, then it breaks, and you end up chasing the store?
well, returnly gives you and the merchant one place to sort it out, with an AI assistant helping along the way
try it: returnly-nu.vercel.app/
Steve (Builder.io) retweeted
spent the week building handoff for the agent-native hackathon
runbooks where me and an agent work the same checklist. some steps are mine, some are its
it drafts, i approve. asked it to approve its own work and it said no lol
built on agent-native by @builderio @Steve8708
Steve (Builder.io) retweeted
Imagine running an online business, making sales, delivering products, attending to customers, and doing everything right.
Then one day, you receive a payment dispute.
A customer claims they didn’t authorize a transaction, didn’t receive what they paid for, or simply disputes a payment you know was legitimate.
Now you’re dealing with something entirely different from running your business.
You have to go through transaction records, search for payment receipts, gather screenshots, find conversations, organize delivery evidence, and put together a convincing response before the deadline.
And here’s the frustrating part: even when you have all the evidence, presenting it properly takes time.
Miss something important, submit incomplete evidence, or fail to respond on time, and you could lose money from a transaction you already fulfilled.
For small businesses, this isn’t just paperwork. It’s lost revenue, wasted hours, and unnecessary stress.
That’s the problem I wanted to tackle.
So I built Parry for the @builderio Agent-Native Hackathon.
Parry is an AI-powered chargeback response assistant built around Paystack.
When a customer disputes a payment, Parry helps gather the available transaction evidence, organizes it, drafts a structured rebuttal, and presents everything for your review.
Instead of spending hours trying to figure out what happened and how to defend the transaction, you get a prepared case and one important decision to make.
Approve it or reject it.
Parry never submits anything without your approval. You’re always in control.
The goal isn’t to replace the merchant’s judgment. It’s to take away the repetitive, time-consuming work that comes with handling payment disputes.
I want to make chargeback management less stressful for businesses, especially smaller merchants who don’t have an entire team dedicated to handling disputes.
The product is live, and I’d really appreciate some real-world feedback.
Demo: clips.agent-native.com/share…
Try Parry: parry-app-topaz.vercel.app
If you run a Paystack-powered store or have experience dealing with chargebacks, I’d love you to test it and tell me what’s missing.
For now, use your Paystack TEST key.
I’m building this to solve a real problem, and feedback from people who actually experience that problem will mean a lot.
Announcing the Agent-Native hackathon!
$10k in prizes. Come build agentic apps with us
builder.io/hackathon
Announcing the Agent-Native hackathon!
$10k in prizes. Come build agentic apps with us
builder.io/hackathon
Steve (Builder.io) retweeted
🔥 $10,000 in prizes. Agent-Native Hackathon. 👾
Most AI agents still hand you a wall of text and make you go dig for what they actually changed.
So we're running a hackathon for agents with actual interfaces. Build an app where the agent has a real job, and people can see its work, edit it, and tell it what to do next.
That's how I run my own task list now. I leave a comment on a task, an agent does the work, and I fix whatever it got wrong right there on the page.
It's free, open worldwide, and you can go solo or bring a team. Submissions close Nov 9. Agent-Native is free and open source, so starting is one command.
Steve (Builder.io) retweeted
Seeing open-slide mentioned in the code is surreal 🤯
@Steve8708 absolutely love this project!! 🩵
agent-native.com/
