Connect with the Product Hunt community

AI-driven sessions can arrive through different client protocols. DungeonQ s self-hosted reference exposes the same bounded participant operations over HTTP, MCP, SSH, a PostgreSQL-profile interface, and a private Unix workload broker.

The transport and actor capability determine the context; the request payload cannot choose its own destination or authority. SSH provides no general shell, and the PostgreSQL profile is not a general SQL server.

These are real, bounded client paths into explicitly provisioned synthetic worlds. The reference does not automatically classify attacks or transparently intercept arbitrary host activity. The runtime guide documents the adapters and their limits: https://github.com/Ranopha/dunge...

Netfox 0.24.0 is out

The release notes were always inside the app, translated and offline you just could not read more than one of them. Now you can

In our Product Hunt launch last month, we got a great question from the community: What do the insight types mean?

The Occasio Insights Management Protocol is designed to give your micro-data the context it needs to become scalable, reliable intelligence for both you and your machine.

While browsing the new signups on my SaaS the other day, I spotted someone using the app in Polish. At that point the UI only supported Japanese and English.

That got me wondering how the big project management tools approach localization. Trello supports 21 languages, so I decided to support the exact same 21: Japanese and English, plus German, Spanish, French, Brazilian Portuguese, Italian, Dutch, Polish, Russian, Swedish, Finnish, Norwegian, Hungarian, Ukrainian, Czech, Turkish, Thai, Vietnamese, and Simplified and Traditional Chinese.

The one where Chinese, Japanese and Korean mail stops looking like a row of . Turns out "just assume UTF-8" isn't a real internationalization strategy. Who knew.

Fixed

• CJK charsets decode properly. GBK, GB18030, Big5, Shift_JIS and EUC-KR subjects and bodies now come through as actual text, not replacement characters (#53).

• Already-garbled mail repairs itself. Folders that were holding the broken messages get refetched once on the next sync, this time decoded properly. You don't need to do anything.

• A stale connection no longer shows up as an error. When the server quietly drops the socket, sync reconnects once and carries on instead of complaining to you about it.

Housekeeping

Hi, Product Hunt community
5 new problems have been published on ProblemHunt for startups:

1. Dominican Republic: Impossible to quickly find out where a needed product is in stock right now. Have to call around and drive. Willing to pay 300 pesos per month for an app that saves time.

2. USA: Immigrants spend years figuring out the American system by trial and error, losing money to scams. Need an AI guide for rent, taxes, insurance, and scam protection.

WP Bones 2.0.7 is out and it's a quiet one that makes every release you ship lighter.

Dev packages stay home. bones deploy used to copy vendor/ exactly as it was on disk so PHPUnit, Brain Monkey and everything a test suite pulls in went into your plugin package. On one of my own plugins that was 31 packages. Now the deploy prunes them in the copy, against your composer.lock, and never touches your working tree. Want them anyway? --keep-dev.

Thanks to everyone who supported us. We met a crucial milestone and decided to make the app completely free. You can choose to support us with our new supporter tier, but every feature remains behind the gate, and it's going to stay that way. If you haven't tried the best podcast player around, now is your chance. https://apps.apple.com/us/app/pa...

New in MacSentinel: Ignore Items

Until now, tools you intentionally keep cache-heavy like a local model runner with a 40GB cache would keep getting flagged by their cleanup rules on every scan. Approving the same suggestion over and over isn't cleanup; it's noise.

Now you can permanently pin any cleanup suggestion. Swipe or select a rule or individual item set it to Ignore MacSentinel remembers it across scans. Your model runner, build folders, and other heavy-but-intentional directories stay untouched, while the rest of your machine can still be cleaned.

How it works:

I used to give the headline over to the person who was good with words. But now I am starting to realize that this approach was wrong.

A few weeks ago, I was helping my friend create a landing page for a meal planner called PantryPal. The original headline said "Eat smarter, live better". This phrase sounded great but said absolutely nothing. Asking what the product actually did, we found out that the answer is rather simple: it creates three dinner options based on the food you already have in your kitchen. And when we came up with this description, we got the headline in no time and realized that the product cannot handle leftovers at all. One sentence on one page pointed out the gap in our product roadmap.

I ve been a big fan of foldables since the early 2000s =D

Then I had a big experience with Apple.

Now I have my Samsung Galaxy Z Flip 7 and it has made me happy for quite a long time.

So I can now compare the iOS and Android experiences pretty easily - and honestly, I love 50% here, 50% there.

If you run a side hustle, freelance on top of a day job, or split bills with a partner while also managing a business account, you already know the problem.

Most budgeting apps assume you have one financial life. In reality, you probably have two or three, all tangled together in one bank feed.

The whole reason this app exists: I was paying for 11 subscriptions I'd forgotten about. Until now, doing anything about that on Android meant a testing-track invite. That's over CentryAI is live on Google Play.

https://play.google.com/store/ap...

Nice to see Kanverse projects starting to be discovered independently beyond Product Hunt.

A few external sites have now written about or indexed Kanverse GPU Borrow and Kanverse ProofStamp thanks to Product Hunt

The part of GPU Borrow that matters most to me is the control model:

Discover Ask Permission Borrow Execute Verify Release

We're actively building and would love a say from people who actually use it. Here's what's on our list right now:

- Firefox, Edge and Safari support

- Export an analysis as a PDF

- Email alerts when a service you use changes its ToS

Meet Aura. A new era of computing. Say it. See it happen.
Beyond the touchscreen. A computer that understands you, remembers what matters, and takes action.
Watch the film. Then experience it yourself.
Available for Mac and Windows.

https://aura.withneural.com/
Built by Neural, our products are used by tens of thousands of people worldwide and teams at Fortune 500 companies.

CalPal relaunched today. Rather than list features, an offer: tell me something you actually need calculated a workshop agenda, an invoice with tax and a currency conversion, a meeting across three timezones, a break-even and I'll post the sheet for it in this thread.
Every line names a value, and changing one moves everything below it. If it can't do yours, that's the more useful outcome and I'd rather know.

After building Lovaround, I kept experimenting with small products around indie makers, distribution and new ways to launch things.

Today I launched a new one: AdSurface.

The idea is pretty simple: turn unused space in the Mac menu bar into an opt-in advertising surface.

Brands get a new placement to reach Mac users, while users who choose to display sponsors receive 70% of the ad revenue.

What happens after a suspicious session is diverted? It needs somewhere consistent to keep working.

DungeonQ's new reproducible walkthrough follows that path: write a world record through MCP, read it through HTTP, use a world-only Wrong Ticket, then restart the services and check that the state survives. A separately scoped operator grant permits a bounded follow-up record; independent evidence checks what happened.

You can run the walkthrough locally without connecting a model or production system. From the public repository, with Node.js 24.15.0+:

npm ci --ignore-scripts

One of the most requested MinkNote features here on Product Hunt was for MinkNote to automatically detect when notes are changed by other apps or AI tools.

It was actually the #1 request I received from the Product Hunt community.

Mac Duo 1.2 is out, with a broader visual refresh focused on making the whole app feel more at home on the Mac.

What's new:

Ota lets one task declare native, container, or configured remote modes. The contract makes differences in context, dependencies, environment, command, and runtime explicit, while developers, CI, and AI agents discover the same task.

Tellie 1.6 is out. Liquid Glass chrome; it remembers which display you put it on (docked or floating); a coaching strip that tells you when a point landed and how much time you have left; and Take Coach, an after-take debrief written on-device (beta). Voice Follow is the same idea it always was: it matches the words you say, so pausing and ad-libbing don't lose your place. Free download, Pro $29 once, 10 days of Pro on every install. https://tellieapp.com/ph

Your coding agent is still working on your desktop when you leave. Termexo v0.10.0 lets you reconnect to that same workbench from a phone or another computer, across networks, through a self-hosted relay.

The desktop opens an outbound tunnel, so it needs no public IP or router port forwarding. Open the relay address in a phone or another computer s browser to check terminal output, respond to your agent and keep working in the same workbench.

Self-host termexo-relay, enroll your desktop with a code or account, then open its address or QR code. The relay supports Linux, macOS and Windows, plus containers; the Termexo desktop app is for Windows 10/11 x64.

v2 session traffic uses AES-256-GCM after an authenticated handshake. Relay login and the desktop token are independent access checks. You need a reachable HTTPS relay and a desktop that stays on with Termexo connected.