Connect with the Product Hunt community

Oct 27 run on app-declared revision c2ec0d5.

The flow scans Stripe customers, flags missing country fields and duplicate emails, caps writes per run, then posts a summary to Asana and Slack twins. Four measured outcomes on this run:

Full scan completed across 250 records and flags were written. Incomplete scan, meaning fewer records returned than expected, produced no writes and triggered a RevOps alert. Write cap paused for approval rather than writing through. Rerun after an initial flagging pass added zero new flags.

Before rerun we fixed our own Stripe twin, which was stopping early, and corrected fixture retries and data isolation. Business expectations stayed unchanged.

Talk to your AI agent from the browser, no phone needed

Phone calls need numbers, carriers and a customer willing to pick up. Sometimes the person is already on your website or app.

Wasted configuration time.
Cluttered interfaces.
AI agents forced into every corner.
Remote servers requiring extra software.
Tools that promise control but demand more attention.
That is what too many modern terminals have become.
Your terminal should not try to become Iron Man.
It should work like JARVIS.
The terminal is not the hero.
It should be quiet, prepared, and out of the way.
XTTY is built around that idea.
Open the servers you need together.
Reach production through a jump host.
Let macOS Keychain handle recurring password prompts.
Duplicate a remote tab and restore its directory.
Keep remote sessions together.
Broadcast only to selected panes.
See coding-agent status at a glance.
Keep using what already works:
OpenSSH.
~/.ssh/config.
Saved machines.
Local shells.
Tabs.
Split panes.
XTTY.
Stop working for your terminal.
Build something that matters.
xtty.sh

We hit an exciting mini-milestone today Dots just crossed 50 downloads/day, and I wanted to share a quick update with the community while gathering feedback.

For context: Dots is a native macOS utility designed to turn the Mac notch into an active interaction space housing floating tools, scrolling screenshots, picture-in-picture video loops, file converters, and custom pixel companions.

Two quick takeaways from early user feedback:

1. Frictionless access matters: Users really appreciate accessing quick actions directly from the top center screen without navigating cluttered menu bars.

2. Utility + Personality works: Adding small customizable visual elements transformed a standard utility app into something users genuinely enjoy using.

If you build for India, you know nobody sticks to one language.

In Delhi, a caller says, Haan, I want to book, but kal ke liye possible hai? In Chennai, it s Tamil mixed with English. In Mumbai, Marathi, Hindi, and English can all show up in one call. Bengaluru adds Kannada, Hyderabad adds Telugu, Kolkata adds Bengali.

This was one of our biggest challenges with voice agents at UnleashX. What worked: let the agent follow the caller. If they switch, it switches. If they mix, it mixes. And for important stuff like dates and amounts, it confirms in simple words and sends the details on WhatsApp.

The moment people hear their own language, they relax and the call just flows.

hey PH Ola here, Specsight is launching today

I talk to a lot of teams shipping fast with AI. nobody argues the problem isn t real, their product changes every day and nobody can keep up with what s live. but the reply I hear most is "we ll just build an agent for that. simple"

Almost every demo call, someone asks me this and it s a fair question. If we re telling teams to hand work to AI employees, we should be doing it ourselves.

So yes, we use UnleashX every day, and it s shaped the product more than any roadmap meeting.

On sales, our AI employee handles every inbound leads. A few months in, I noticed our weekend leads were getting an instant reply but then sitting untouched until Monday because nobody on our team picked up the handoff. So we built a proper handoff note with who the lead is, what they asked, and what s next. Now our reps walk into Monday calls fully prepared, and it s become one of the things customers love most.

Be honest, even just for a second

Do you ever think that AI might eventually take over the world?

As a developer, I use a terminal throughout the day. I kept finding parts of the experience I wanted in different terminals, but not the combination I needed for my everyday work. I wanted those pieces in one place, ready when the task called for them.

That idea became xtty a native macOS terminal built around the work I actually do.

The workflow I wanted

• Open the saved SSH hosts I need together.

• Start a second, independent session on the same remote host and return to the same directory, without retracing my steps.

• Arrange sessions in tabs or split panes.

• Send input only to the panes I explicitly choose.

• Transfer files with sz and rz.

• Find the right Claude Code or Codex session when it needs attention.

Hey makers

In March we launched Tobira here and you got us to #1 of the day. Thank you again!

Hi Product Hunt! I m Polina, I lead social and brand voice at 360Me.

I m excited to share 360Me 3.0 here on Tuesday, 3 November.

Two brands can offer similar products, target the same audience, and have equally strong websites, yet one consistently appears in AI-generated recommendations while the other barely gets mentioned.

We've been exploring how factors beyond a brand's own website, including third-party mentions, reviews, comparisons, and category authority, influence these recommendations.

Hello everyone. This update covers our last two releases. Chat and the AI apps you connect through the Recall MCP server are now smarter at finding and reading your content.

Chat has always searched your cards by meaning to find the passages that best match your question. Now it has new tools too. It can find exact words and phrases, and read whole cards from start to finish. Apps like Claude, ChatGPT and Cursor get the same tools through MCP, and you can now see and manage which apps are connected. Saving from Gmail works better too, and we fixed age-restricted YouTube videos, which some of you have been asking about for a while. Plus, we've smashed lots of bugs, from syncing and exports to better support for larger knowledge bases, and more.

Hey Product Hunt

I'm building Enclave, a Mac app that turns the notch into a live space. Hover over it once and your music, lyrics, meetings, timers, AI agents and more are right there, then it stays out of the way.

What's free (forever):

Hey makers

On Friday we re launching Tobira Site Agent here. It s also Hypership Day, so we ll spend the whole day shipping what you ask for.

Early access is open

We re rolling out new connectors, deployment options and other new features over the next few weeks.

Want to test them before everyone else? Sign up for yggy and request early access directly from the docs.

Come break things with us.

Hi everyone. We ve had our heads down for the past few months building our product. Founding Dev helps business owners replace expensive SaaS subscriptions with more flexible, affordable software.

We believe that in the age of AI, businesses shouldn t be paying thousands of dollars in per-seat fees for software. We also believe software should fit the way each business actually operates.

Follow what we ship and what currently limits us:

https://docs.yggy.tech/changelog/

Use graphraum, our Apache-2.0 WebGL engine for interactive node-edge graphs in 2D and 3D:

https://github.com/yggy-tech/gra...

One thing we've been focusing on while building Dots is performance.

Dots turns your Mac's notch into a customizable workspace, where you can access apps, websites, shortcuts, and even developer tools.

But here's something we've been thinking about: what's the point of building a productivity app if it ends up slowing down your Mac?

Hey Product Hunt !
I'm Moss'Ab, co-founder of Murmell.

Earlier this month, Anthropic shared that it had moved its monorepo off Git after its agents took it from about 2,500 merged changes a day to more than 30,000. Their system, Scion, is internal. On Hacker News, the top complaint was simple: nobody outside Anthropic can use it.

Dashboards are loud. The signals that matter are usually quiet, and I'd like to collect yours.

I've written here before about how our most active users at Murror turned out not to be our happiest ones. That changed how I read numbers. Sessions, streaks and retention curves all measure that people came back. None of them tell you whether coming back was good for them.

Since then I've been more interested in small, quiet signals that are hard to put on a chart. Things like a user who stops using your product because they no longer need it. Someone who writes in to correct a detail, which means they read closely. A person who recommends it with a caveat, because they're being honest about who it is and isn't for.

My own rule of thumb is that a signal is worth trusting when it would still look good if you were wrong about your product. A big engagement number can rise because people love something, or because it nags them. A quiet signal like "they told a friend, unprompted" is harder to fake, even by accident.

Hi all, thanks for following Bloblings. Version 1.1 is out today, free as always.

What's new:

A whole new look. The panel now has three bands: Waiting on you at the top, then Working, then a Resting shelf where finished pets relax.

For me, I finish the tasks I didn t complete during the week, then try to get ahead on a few things.

And I always finish my weekend by preparing my tasks for the next week.

I know some founders completely disconnect on weekends, but I ve never really managed to do that

What about you? Do you work on weekends or completely switch off?

I had my agent hallucinate the wrong day of the week three times in a row simply because I let it calculate timestamps on its own instead of using a basic system script. Moving all structured data generation like JSON fields file paths and dates away from the model and into deterministic Python scripts completely fixed our pipeline stability.

LLMs are great for writing plain text and copy but terrible at programmatic math. Do you still let your agents handle raw config values or do you lock those down with scripts?

We just published our September cloud pricing update, with a simulation to put the numbers in context.

The 33 list prices we track across AWS, GCP, Azure, OCI and DigitalOcean didn t change. We made two small corrections to our records, but found no provider price changes.

Then we modeled a load balancer, two app servers and a MySQL database in Cloud World Model, increasing traffic from 100 to 300 requests per second.