PH热榜 | 2026-06-13

I remember using this drag‑and‑drop style deploy on Netlify years ago, so Vercel definitely took their time bringing Drop out.

Netlify’s new credit‑based pricing (where production deploys cost credits) actually makes this Vercel feature look pretty reasonable now.

The no Git/CLI/local setup part is the real hook for me. I can see this being useful for quick repros or static preview handoffs. Curious how Drop handles teams that need preview env vars or a small build step instead of just plain HTML.

INCREDIBLE! You guys just saved my relationship and my sanity. 😭

Ever since I showed my girl Claude Design, apparently every thought she has needs a production link.

A grocery list.

A checklist.

A note to herself.

One time she recreated our text messages and made me deploy them so I could 'review my mistakes.'

Thank you for enabling this insanity. 😂

The app isn't bad, but a short introductory video would have been better. Also, if we drag and drop a program file, will it be able to run that file just like it would on a server? In other words, if we share a URL with people, will they be able to edit the file at that URL—just like a Google Doc?

Great timing for this launch. How does the CDN performance compare to the previous version?

“This is such a smooth idea. Sometimes you just want to get something live without touching Git or a CLI, and Drop nails that. I’ve been building my own small tools lately and anything that removes friction from shipping is a win. Drag → deploy → share in seconds feels like the right direction.”

This is so Vercel. No setup, no CLI, just drop and it's live. I already use Vercel for my own app but honestly I'd use this just for quick demos and sharing WIP stuff with testers. Does it support environment variables or is it purely static for now?

Awesome and congratulations! It is only for html files?

This is one of those ideas that sounds obvious after you see it. I've definitely had moments where setting up a deployment felt harder than the actual project

Curious if Drop is mainly meant for simple/static projects, or if you see it becoming a bigger onboarding path for people who later connect Git and continue building from there?

Treating the browser as the deployment interface is a nice inversion. The File API already abstracts the filesystem correctly, so there's no reason to force a Git remote into a workflow that's just 'I have files, make them live.' We've hit that friction with static prototyping. How does Vercel handle cache busting when you re-drop to the same project name?

This is slick. Does Drop support any framework detection, or is it purely for static files?

Hi everyone!

Kimi K2.7 Code is open-weights and focuses on improving real-world long-horizon coding performance. Compared with K2.6, it shows clear gains in instruction following over long contexts and higher success rates on multi-step coding tasks.

It also reduces overthinking quite a bit, with 30% lower reasoning-token usage. The model runs with thinking mode on by default and has better support for vision + tool calling in agent workflows.

Kimi Code has already upgraded its default model to K2.7 Code, and a 6x faster high-speed version is coming!

The 30% drop in reasoning tokens alongside better multi-step task success is the interesting signal here. It suggests you're pruning unproductive reasoning chains rather than just thinking less. We've seen agent costs spiral on complex multi-turn tasks because of runaway chain-of-thought. How did you train the model to distinguish productive reasoning steps from redundant ones?

Congrats on today's launch!!

The open-weights + 256K context combination is what I'd test first, especially on a repo task where the model has to keep tool outputs, diffs, and failed test logs straight. Lower reasoning-token usage is useful, but the tradeoff I wonder about is recovery after the agent makes a bad edit. Do you have evals that measure whether K2.7 can backtrack from a failed test run without losing the original instruction?

Interesting launch. For coding-focused models, the thing I’d want to test is not just generation quality, but how well it handles long-running repo work: keeping context clean, explaining risky changes, and recovering after failed tests.

Interesting model. The 30% lower reasoning-token count is notable. Does that also reduce latency proportionally for typical multi-step tasks?

To be honest, I really like Kimi, but this time the benchmarks are a bit below my expectations; they only seem to be slightly better than 2.6. But I really appreciate the fact that you’re open-source and constantly striving to improve. Thanks, team.

The automatic maintenance angle is clever. Having the agent detect and regenerate scraping code as page structures shift solves one of the biggest pain points in web data pipelines. We've spent significant time dealing with brittle scrapers that break silently and cause data quality issues downstream. How does Prometheus decide when a structural change warrants code regeneration vs. treating it as a transient anomaly?

Web scraping for agents is one of those problems that sounds simple until you actually try it. Curious how it handles sites with heavy JS rendering or login walls — that's usually where these tools fall apart. Will test it on a few of my usual sources.

If you host it, can you see exactly what code it generated and edit it when you need to?

the run it yourself or let us host split is interesting because those are pretty different value propositions. running it yourself means you're still responsible for maintenance even if the agent writes the code. hosting means you're trusting Firecrawl to maintain the extraction logic as sites change, which is a significant operational commitment to make on someone else's behalf. curious what the SLA looks like for hosted maintenance and what happens when a site change breaks extraction and you need the data today

This is neat. When page structures change, does Prometheus auto-detect the break or wait for you to flag it?

The maintenance part is what makes this useful. Writing a scraper once is easy enough, but keeping it working as pages change is where most web data projects quietly become a time sink.

The forward-deployed agent framing is useful for web data, because the hard part is usually maintaining extraction when pages change. Curious how you decide what should become a reusable workflow versus a one-off crawl for a customer?

Prometheus generating the collector code is cool, but the part I’d love to see is whether it also generates validation checks with it. For example, after it samples the data, can it create simple assertions like required fields, expected value ranges, stable selectors, or row counts so teams can catch bad runs before the data reaches downstream systems?

Stumbled across this morning and let me tell you in just an hour of using it I have extracted immense value when it comes to building websites for local businesses and extracting specific components. I have been using the Firecrawl plugin in Claude Code for about 3 months now and to have a streamlined way to use it separately has been very helpful so far, cheers lads.

no accounts, no ads, no data collection for a kids app is the thing that should be the default and almost never is. the fact that it runs on-device means there's no server to breach and no data to sell. that combination alone separates this from most kids education apps and it's worth leading with more prominently

Congrats onn launching an interesting product! The on-device constraint probably looks to the main product decision here? Most people would've made a cloud call and gated snaps behind a paywall. What if a kid points at a mug, the model could land on cup, mug, or the wrong word in the target language. Who wins that call, and does the kid get to correct it?

I like this observation from kids.

How does this system actually let me "point" though? Do I learn through a little popup or browsing through things I've poitned at, on the app?

Very cool. Love on device AI work. Im curious, what was the toughest part of getting this to run on-device?

I had this on my mind and am happy that finally someone made it! Especially for the youngest ones :)

The Word Dex idea is a nice touch. It turns language learning into a little house-wide scavenger hunt instead of another flashcard app, and on-device processing feels especially important for a kids app.

Love the on-device approach – no accounts, no data collection is a real differentiator for a kids app. My son would've used this. Congrats on the launch!

Hey everyone, I’m Matt 👋

I’ve lived in a few states over the years, and whenever friends visit those states they always ask where they should eat.

That got me thinking… if you're headed to San Francisco and have friends that live there, they probably know the good spots better than a random review.

So I built NomNak to help people find restaurants through people they trust, save spots to try, and build their food passport.

Fun fact: My wife likes to make up words. When food is really good she calls it a “nommy nack” (basically a yummy snack). I shortened it to NomNak so people could actually spell it :)

The app is a passion project and completely free to use. If you try it and enjoy it, sharing it with your foodie friends and giving NomNak a follow would mean a lot ❤️

It's true. Google maps is truely just mid...

Android or webpage would be great for NomNak

why didnt you release this earlier, ive been waiting for something like this

Just wanted to jump on here and say thanks for checking out NomNak and sharing feedback. Really appreciate all of you ❤️

Love food related plays - how do you compare yourself to Beli?

positioning of seeing where friends actually eat versus where they recommend is an interesting distinction. recommendations require effort and create social pressure to only suggest places you're confident about. passive check-ins or logs require less commitment and might give a more honest picture of someone's actual eating habits. which one is the core mechanic, active recommendations from friends or passive tracking of where they go, because those create very different social dynamics on the app

Interesting one! The visiting-a-new-city case is both the strongest pitch here and also the hardest to actually pull off? The whole thing leans on a trust graph but the moment people need it most (landing somewhere I dont live) is exactly where I'd have the fewest friends on the map

Restaurant recs from friends are usually way better than scrolling through random reviews. The Food Passport idea also gives it a nice personal history layer, not just another saved places list.

Nice man do you plan to have android version?

it's free and no ads? Love a good passion project, will definitely share this around with my foodie group chats.

Separating the avatar entry point from the broader creative suite is a good product call. Talking-head synthesis has its own latency and quality tradeoffs that get lost when bundled into a general media workflow. We've looked at async video for customer communications and lip sync accuracy is consistently the bottleneck. How do you handle audio-visual sync drift over longer clips where accumulated timing errors become visible?

This looks interesting. Can you bring your own avatar or are you limited to ElevenLabs' stock ones?

Congrats on the launch!!!

Is this going more towards the "social impact" side or the "saas product" side? Im really interested in its direction

Hey Product Hunt, Fiza here, co-founder of Feezza.

I spent three weeks logging every meal perfectly, hit every target, drank the water, got the steps and still woke up exhausted with no idea why. My health app said my macros were perfect. That question then what is the point of all this changed everything.

Sukhraaj and I built Fiza because she didn't exist. He had never written a line of code. I had never built a product. What followed was the hardest year of our lives.

Fiza is not a feature list. She is a daily health relationship. She learns what you eat, tracks your conditions, detects hidden calories from how food was cooked, checks your medications, and quietly connects patterns in your body you have never seen before.

Today she is live on the App Store. We would love your support, your feedback, and your honest questions. Ask us anything.

This is a great product and initiative to meet everybody's goals inclusive of their health conditions. I have been tired of meeting unreal numbers and following a strict unnecessary program that affected my mental health in the long run. Can't wait to better use my diet and mobility to aim for a good health as goal rather than just a weight loss or loosing nutrients.

How does it estimate things like oil used or portion size without guessing too much?

The part that feels useful is looking for patterns over weeks, not just logging today’s meals. A health app that helps explain why you feel a certain way is much more valuable than another tracker full of numbers.

Hey Product Hunt 👮
I built ScrollJail because I had this exact problem.
I deleted Instagram my brain switched to YouTube Shorts. I deleted YouTube: I ended up doomscrolling on Snapchat. That's when I realized: the app was never the problem. It's the three-second window where your thumb opens something before your brain knows what's happening.

What ScrollJail does:
→ Every time you open a blacklisted app, an AI asks: Why are you here?
→ Good reason: you're in, timer starts
→ Bad reason: the Warden gets creative
→ No "Ignore Limit" button. No bypass. One question you can't skip.

What makes it different:
→ App-agnostic: blocks the behavior, not the app
→ Free core experience, no paywall before value
→ The AI has personality and users are already sharing their "Roast Receipts" unprompted

It's currently a free iOS Shortcut. 3 minutes to set up. Native app is next if this gets traction.

I'd love honest feedback: does this feel like a real behavior change tool to you, or a novelty that wears off after a week? That's genuinely my biggest open question.

From Moritz, the guy building ScrollJail

P.S. 20% off all launch weekend. Use code PH20OFF at checkout.

So proud of what the team pulled off here. LFG 🚀🚀🚀

Wow!!

Do you have a quick way to move intensity up or down mid session without rebuilding the whole mix?

Do you have clean platform specific fields like crossplay, controller support, and Steam Deck compatibility?

I like the idea, good launch :)

You could maybe add a coin system so that users can see the other 3 un-flipped cards. and they have to do challenges or P2W for coins lol

The limited daily discovery is a good constraint. Most social apps make you keep scrolling, so a slower card-based flow feels better suited for actual conversations instead of just performing for attention.

Now also available from the Shopify app store: https://apps.shopify.com/praisepage

This hits close to home. Watched too many organizers burn out juggling Google Forms for speakers, WhatsApp groups for announcements, and three different ticketing logins all for a single event.

What stands out:

• The workspace model is smart. Most platforms assume you run one community. Event Parlour gets that the best organizers are running multiple simultaneously.

• M-Pesa is already live. Not "coming soon" live. For the Kenyan market, that's not a nice-to-have, it's the difference between an event selling out and stalling at checkout.

• The Storefront is a quiet game-changer. Organizers can now monetize beyond tickets merch, resources, whatever fits their community.

• And the Hangouts page? Finally a platform that acknowledges not every gathering needs a formal ticketing flow. Informal meetups deserve proper infrastructure too.

• Analytics , we have web analytics that tracks how many people checked on your listed event and where they are based and an add on we have announcements , where you can announce changes to your attendees.

• For attendees , track events you attended , book events and transfer tickets incase you had an emergency , ticket resell are coming soon.

  And if you're an organizer ready to get verified and start hosting KYC is live: https://app.eventparlour.com/pitch-deck

Also what's the biggest challenge you've faced on other event platforms? Scattered tools? Hidden fees? Payments that take forever to land? Drop it below