PH热榜 | 2025-12-04

👋 Hey everyone, I'm Hoshang, Co-founder of Pylar.

Super excited to finally share what we’ve been building.

Agents today are great at reading docs, invoices, websites, transcripts -
but the moment you want them touching structured systems where sensitive customer data is stored e.g Snowflake, Postgres, CRMs… things get tricky.

We kept hearing the same two blockers over and over:

• Agents may over-query and silently spike warehouse bills

• Agents are at a risk of leaking sensitive data (PII, financials, customer history) because access isn’t properly scoped

And right now, teams have two options:

- Off-the-shelf MCP servers : 18,000 exist, ~10% are malicious, and most are exploitable or too generic for production.
- Custom API wrappers : months of engineering bandwidth used up in building endpoints, policies, and governance… all brittle, fragmented, and hard to audit.

This forces companies into a painful choice: lock agents down so much they become useless, or open things up and risk a security incident.

Traditional database ACLs weren’t designed for autonomous systems. Custom APIs are hard to build, govern and control for agent level interactions.

Pylar exists to fix this. It’s a governed access layer between your agents and your entire data stack.

You connect your datasources → define sandboxed SQL views → turn them into MCP tools → ship them to any agent builder… all from one control plane, with full observability.

What you get out of the box:

• Agent-specific sandboxed views (never raw DB access)

• Enforced permissions & guardrails

• Automatic breach containment + audit logs

• Publish to any agent builder (n8n, Cursor, Claude, LangGraph, etc.) via a single secure link

We’re already working with some fantastic data, platform, and security teams - everything from internal analytics copilots to customer-facing AI features wired directly into production data.

If you’re exploring structured-data access for agents, I’d love to hear your thoughts, help you build your use case or just share best practices on what we've been seeing with our customers. You can book a call with me here if you'd like.

Thanks for checking us out — means a lot. 🚀

- Hoshang
Co-founder, Pylar

Finally, someone tackling the agent-to-DB mess. I’ve nursed a painful Snowflake bill from a runaway agent. Sandboxed views + audit logs feels sane. How do you cap query spend per agent? Might wire this into Cursor first, then LangGraph if it holds up.

Congratulations on the launch 🎉 🎉 !!

@vishalbajaj Great product! All the best for your launch 🎉

Does Pylar throttle or rate-limit agent queries in any way? Congrats on the launch.

Congrats on the launch @hoshang_m this solves a serious gap for teams working with sensitive, structured data. I’m curious how Pylar handles evolving permission needs over time. If schema or data-access policies change, can sandboxes and guardrails adapt without teams rebuilding the entire setup?

This is pretty cool, I have been working with some fintech companies and the sheer volume of data they have is ginormous.

I love how it abstracts the base layer queries and have complex queries converted to tools. amazing job guys! Will try it for sure

@Pylar Congrats on the launch! Securely connecting entire data stacks to AI agents is exactly what teams need as AI adoption scales.

Data security and access control are critical when agents interact with sensitive information. How does Pylar handle permission management across different data sources?

Are developers able to set granular access rules for specific agents or use cases? Curious about how the authentication flow works when connecting multiple platforms.

I could see how integrating Pylar’s AI-driven workflow automation could streamline content and data processes inside our tool, and I’ll go check it out to explore the possibilities.

It's an amazing idea. So can the agent run analytical queries in the DB as well?

congrats on the launch!

amazing product guys well done !!

@hoshang_m Congratulations. And happy product launch.

Congrats Hoshang! Pylar seems like a huge step forward for safely connecting agents to structured data.

How do you monitor agent behavior across different builders (Cursor, LangGraph, n8n, etc.) from one place?

Custom APIs are a nightmare and MCP servers are a minefield. This feels like the first real governed layer built for agents. Good work!

Don't fully understand what this is. What's the top 3 use cases for this?

This tool caters to teams building or deploying AI agents: it lets agents leverage internal data (to deliver context-rich outputs) while maintaining strict control over data access—addressing both efficiency and risk management needs.

Do you plan to expand compatibility to niche or industry-specific datasources (e.g., healthcare EHR databases, financial ledger systems) for specialized use cases? Also, will Pylar include pre-built access control templates aligned with common compliance frameworks (e.g., HIPAA, GDPR) to accelerate secure setup for regulated industries?

Great launch. congrats. Do you have custom guardrails option too?

Congratz on the launch team! With agentic AI this will be a must.

congratulations for your product!!

Love what Pylar is solving. As someone who works with founders handling fast growth + ops, I know how often “data safety + speed” becomes the invisible pressure behind the scenes.

Awesome! Is it possible to have an agent that sync my Notion page with notes of different customers to my CRM system Attio?

Hi PH! 👋

We’re the makers of GNGM — a tiny habit for big sleep wins. After struggling with irregular sleep and feeling overwhelmed by trackers and metrics, we built something deliberately simple: one short, calming check‑in each night that helps night‑owls rebuild a consistent rhythm without pressure or data noise.

Why GNGM?

• No trackers, no wearables: nothing to sync or monitor.

• No shame or strict rules: a gentle routine, not a regime.

• Designed for night‑owls: realistic prompts and timing that respect your lifestyle.

• Minimal, calming feedback so you can see progress without getting obsessed.

How it works

1. Quick nightly check‑in

2. Subtle cues and nudges to help you wind down and establish consistency.

3. Lightweight progress signals over time — enough insight to keep you motivated, never overwhelming.

We built GNGM because small, consistent rituals beat massive overnight changes. It’s for people who want to feel better rested without turning sleep into a full‑time project.

We’d love for you to try it and tell us what you think — what helped, what felt off, and any features you’d actually use. Ask us anything below; we’re excited to hear from other night‑owls and sleep-curious folks.

Thanks for checking us out!

— The GNGM team

This launch feels different for me tbh. I pulled way too many late nights building GNGM and kind of… burned myself into making a sleep app 😂

Pretty happy I don’t have to stay up that late anymore. Hope it helps you sleep better than I did while coding it.

Looks polished! Does it support "free-running" sleep rhythm?

Congrats on the launch

Love the minimalist approach! Finally, a sleep app that doesn’t overwhelm with data. Subtle, soothing, and made for night owls — I’m in. 🌙

Do the nightly prompts adapt over time based on my habits, or stay consistent?

Wow using Suno for sleep sounds was super smart. Looks like it’ll be the norm for all kinds of use cases like this. Congrats on the launch!

Congratulations on the launch 🎉 🎉 !!!!

Finally see a product like this! Sleeping is normal, but good sleeping is a big challenge. Hope this app can solve my problem

Love the “no pressure, no trackers” approach. I’m a night-owl who’s failed at every strict sleep routine, so this feels really refreshing. Downloading now.

@jotzilla @polman_trudo Nice approach to sleep improvement without overwhelming users. I’m curious how GNGM measures progress in a meaningful way without relying on trackers, what signals or habits indicate someone is actually improving their sleep consistency?

@justin2025 Congrats on the launch 🚀

Can you please tell me how GNGM breaks our late night habits ?

Do you plan to add optional, low-intensity insights (e.g., mood/energy tracking) for users who want gentle feedback without data overload? Also, will the app include personalized, gradual bedtime shift recommendations tailored to individual night-owl patterns?

How exactly does this software help night owls?

The consistency and rituals actually work better! Really like the idea!

Congos on the launch @justin2025

Dowloaded on android, see if you add setting to enable grayscale mode before bedtime

that will help me as a user.

What a great tool! Congratulations. Sleep is so important and most of we ignore it.

@justin2025 The sleep habit app for night owls is such a practical concept! Resetting sleep schedules is incredibly difficult, especially for people with naturally late chronotypes.

Love that this focuses on habit formation rather than just tracking. How does GNGM help users gradually shift their sleep schedule without disrupting their daily routines?

Are there specific techniques or nudges built in to make the transition smoother? The big sleep wins idea resonates!

congrats on the launch! love the design

Really love the vibe and mission behind GNGM. As someone who naturally leans toward late nights, the idea of a gentle, no-pressure nightly check-in instead of data-heavy trackers feels refreshing — it’s exactly what many “night-owls” like me need. The simplicity (no wearables, no complicated charts) is a strength: just a calming routine to help reset your rhythm. Looking forward to seeing how consistent use can reshape my sleep habits.

Do you plan to add customizable reminders or optional “wind-down rituals” (e.g. light stretching, breathing prompts, soft audio) to make the nightly check-in even more helpful?

looks cool, congrats for the team!

I could see how our tool could plug into GNGM’s streamlined no-code workspace to simplify automation around our community workflows, so I’m going to take a closer look at this launch and explore the synergy.

I seriously need this app! havent slept well since building my current project, and hope GNGM could save me out...

@justin2025 Congratulations. And happy product launch.

So the name GNGM means "good night, good morning"?

Do you plan do add a feature like sharing the sleeping time with family?

Night owl here. Kinda burned out on rings/graphs. One simple check‑in sounds… doable. Curious what the wind‑down nudge feels like—like a tiny prompt or more of a ritual? Either way, I could use gentler evenings. Saving to try this week.

Congrats on the launch! 👏
I haven’t tried GNGM yet, but the concept immediately stood out to me. Most sleep apps feel heavy with data, graphs, and pressure but this focus on a simple nightly check-in and gentle cues sounds much more realistic for people who just want a calmer routine.

I really like the idea of helping night-owls reset without turning it into another “project.” The minimal, non-judgmental approach feels refreshing.

A few questions out of curiosity:

1. How did you validate that a single nightly check-in is enough to build consistency?

2. Do the prompts adapt over time, or are they intentionally kept static to avoid overwhelm?

3. Are you planning integrations (Apple Health, reminders, etc.) or deliberately staying “no-tracking”?

Looks great - wishing you a smooth launch day! 🚀

Really nice launch, GNGM team. From a clarity & onboarding lens: when a user opens the app for the first time, what’s the one belief you want them to leave with in the first 10-15 seconds?
Is it:
• “I’m finally doing a simple routine I can stick with.”
Or:
• “This app understands my late-night rhythm and doesn’t judge it.”
Because in habit-change tools, the biggest barrier isn’t features—it’s the user’s belief that change is possible without pressure. Curious how you’re shaping that.

Hey Product Hunt! 👋

We're excited to launch Documentation.AI, a platform that makes it incredibly easy to build, maintain, and scale beautiful, AI-ready product documentation.

I'm Roop, co-founder of Documentation.AI. Over the past decade, my team and I have built products across video streaming, edtech, AI agents, and more. No matter the industry, we kept seeing the same issue: documentation was always deprioritized, outdated, or both.

But with the rise of AI, documentation has never been more critical. Tools like ChatGPT, Claude, Cursor, and other AI agents are becoming the primary way users discover, understand, and integrate products, and they rely entirely on your docs as their knowledge source. Your documentation isn’t just a reference anymore; it’s a growth channel.

We noticed three big problems teams face:

1. Creating great docs is hard — designing, writing, and publishing takes forever.

2. Keeping them current is even harder — docs go stale the moment you ship new features.

3. Most docs aren’t built for AI — despite AI agents depending on them, they’re not structured for semantic search, embeddings, or MCP workflows.

What makes us different

✨ Ship in under 5 minutes — Beautiful, fast docs out of the box, no design work required.

🔄 Stay up-to-date effortlessly — Our built-in AI agent helps you write and maintain content. It can analyze Git commits, support tickets, and more.

🤖 AI-ready by default — Structured for ChatGPT, Claude, and coding agents with MCP endpoints and semantic markup. Your docs show up where your users actually work.

🛠 Flexible editing workflows — Use our web editor, AI agent, or a code editor/coding agent via a docs-as-code workflow.

💬 Built-in AI assistant — Users get instant, cited answers without leaving your docs, dramatically reducing support tickets.

We'd love your feedback on:

• What's your biggest documentation pain point today?

• How do you keep your docs in sync with product/API changes?

• What would make a tool like this indispensable for your team?

Happy to answer any questions! 🚀

You can also reach us via Slack if you want to chat directly.

Congrats! Do we ask the AI agent for updates or it automatically proactively suggests updates?

Congrats on shipping! The built-in AI assistant for users is super underrated; support teams will love this. :D

Great work!! Any plans to support multi-language documentation with automatic translations or it already supports?

Does your product access my Git and, based on that, generate or edit documentation itself?

Looks good, love the design.

Creates a public GH repo though be aware if you want your documentation to be private.

The integration with Jira/Linear is super helpful. Devs and technical writers would love it.

Love the positioning. With AI agents everywhere, good documentation is becoming infrastructure, not a nice-to-have, but a must-have.

The looks great.
Definitely trying it out soon for for products.

How you handle versioning across major releases? BTW, congrats on the launch

Love the design quality. Fast-loading docs with semantic structure feels like the future of technical writing.

Interesting approach with llms.txt. How customizable is the schema for complex API surfaces?

Congrats on the launch @roopreddy and Team! This resonates hard. We've struggled with keeping our API docs in sync with rapid feature releases for ResumeUp.AI. The Git commit analysis for auto-updates sounds promising.

I love that you treat docs as a growth channel, Roop. That mindset is becoming so important with AI. :)

Congratulations on the launch 🎉 🎉

The MCP integration caught my eye. Does it work with self-hosted LLM setups as well?

The toughest is to deciding what actually needs documenting versus what’s just noise. Products evolve every week, but not every update deserves a rewrite. Finding that balance between being comprehensive and being clear is honestly the struggle.

Product teams always struggle with outdated API docs. The monitoring of support tickets is a brilliant add-on :D

Really awesome. Is there a provision in the generated docs to handle SEO, SEMs, LLM listings, etc., properly?

Hey Product Hunt, I’m Pete, CEO at Dagster! We just launched Compass: collaborative AI-powered insights in Slack powered by your data warehouse. Finally, you can get to the “why” behind your metrics in seconds, not days.

We originally built Compass as an internal tool to solve our own business problems. We have a great data team, and built out a comprehensive data warehouse containing all of our critical business data like sales opportunities, marketing campaign performance, and product engagement analytics.

However, we struggled to really leverage this data at our organization. Our BI dashboards were great at telling us the “what” - how our sales pipeline is trending, how many DAUs we have vs MAUs, what our AWS spend was last week, etc - but it didn’t tell us the “why.” Why is pipeline down? Why did DAUs spike? What drove our AWS spend increase? Etc.

We had these sorts of questions on a daily basis. And every time we did, we’d file a ticket for our excellent  data or revops teams, interrupt their day, and wait a few hours or day or two to get an answer.

Like many companies, we built an internal Slackbot to solve this problem. Anyone could ask it a question in natural language, and it would return an answer, complete with data visualizations and its methodology.

It took off like wildfire, and a majority of our employees started using it on a weekly basis. We decided to turn this internal tool into a fully supported product offering we’re calling Compass.

How Compass is different

We took a different approach than a lot of other AI tools in this space.

🖥️Slack native. We exist solely within Slack. There’s no separate web app to long into, so it’s easy for users to get started immediately.

👥Fundamentally multiplayer. We let data people, business stakeholders, and AI all work together seamlessly to iterate to the right answer.

🏋️Proactive. In addition to answering questions, Compass will deliver personalized data insights every day, surfacing useful trends and starting points for analysis. One of our customers (a highly successful tech unicorn) found a support issue impacting $10mm+ in revenue in the first week.

🧠Crowdsourced business context. This is our secret sauce: we automatically and continuously learn business context based on properties of the data, characteristics of the business, and, most importantly, end-user interactions.

All of this is governed by a gitops workflow, which keeps the data team in control and brings a real SDLC to context engineering.

Who is it for?

Compass is about bridging the data team and the business. Data teams - anyone that manages a modern data warehouse like Snowflake, Databricks, BigQuery or Athena - will benefit, as will their stakeholders: sales leaders, marketing leaders, recruiting leaders, executives, as well as ICs.

Get started at compass.dagster.io. Onboarding only takes a few minutes, and if you sign up before the end of the year your first month is free!

If you dont mind me asking? What are the 3 top use-cases for this?

Strong launch Pete. Compass feels powerful and focused. Congrats to you and the team.

This is the kind of thing that makes you wonder why every data tool isn’t just… in chat already. Nice launch!

How does GitOps-backed context actually work in practice? are all changes versioned and reviewable before going live? Congrats.

Congrats on launch! Dagster Cloud looks super clean.
What’s the first “oh wow” moment you want new users to experience when they jump in?

How does Compass handle ambiguous or poorly phrased queries in Slack? Does it ask clarifying questions or make assumptions?

Do you plan to expand Compass to other collaboration platforms (e.g., Microsoft Teams) for teams using alternative workplace tools? Also, will Dagster offer pre-built connectors for top data warehouses (e.g., Snowflake, BigQuery) to accelerate onboarding for new users?

@Dagster Asking Slack about your data and getting instant answers is a game changer for productivity! Chat based data queries eliminate the need to context switch between tools.

How does Compass handle complex queries that might require joins or aggregations across multiple data sources?

Are teams able to customize the natural language understanding for industry specific terminology? The real time aspect in Slack is brilliant for keeping workflows seamless.

I could see how our tool could benefit from Dagster’s orchestrated, reliable data workflows to streamline automation and analytics — I’ll go check out this launch and dig in further.

How do you handle conflicting interpretations from multiple stakeholders asking questions in Slack?

Absolutely amazing!

Dashboards tell you what, Compass finally tells you why—right where teams already live in Slack. Feels like a superpower for anyone tired of pinging RevOps for ‘just one more chart’.

Hey folks!

We built Protaigé because we saw marketing teams facing a difficult choice: sacrifice brand consistency to move fast, or slow down to maintain quality. We wanted to build a third option: an AI system that understands your brand as well as you do.

Protaigé is a creative agency in the cloud. It takes you from a simple brief to a full campaign in minutes.

Here is what makes us different:

🚀 End-to-End Production: Protaigé automates the entire campaign production process. From your initial brief to the final export, we build every asset in one connected workflow. Unlike many other AI tools that are point solutions and handle just one task like copywriting or banner generation.

🧬 Brand DNA: Generic AI output often sounds soulless. We allow you to upload your brand guidelines and other relevant brand assets and documents. Our system ingests this to ensure every output aligns with your voice and visual identity.

💡 Strategy First: Most tools skip straight to execution. Protaigé starts with strategy by completing the marketing brief and then developing distinct creative concepts for you to review and approve before it builds a single asset.

Who is this for? We designed this for marketing teams and founders who need to scale output without scaling headcount. Whether you need a tactical push or an integrated campaign, Protaigé handles the heavy lifting.

Our Launch Offer: We are currently in Public Beta.
- Campaign Lite: You can try the platform on our free plan.
- Campaign Plus: For this launch, we offer a 75% discount on our Plus plan ($49/month) for small teams.

We want your feedback. Specifically, does the Brand DNA capture your tone correctly?

Excited to hear your thoughts! 👇

Great! Do you also manage social media and email campaigns from your dashboard?

It looks so easy to create a campaign! Great tool, my congrats!
And I was the 200th person to upvote Protaige :D

Seeing a full campaign appear instead of random headlines feels refreshing. No patching thigs together anymore.

Very smooth process!

Seeing a full campaign appear instead of random headlines feels refreshing. No patching things together anymore.

Really like the strategy first approach here. Most AI tools jump straight into spitting out assets, but campaigns only work when the thinking is solid. The Brand DNA angle is also interesting. Generic AI copy is a real issue when you’re trying to protect a brand's voice. Excited to see how well it adapts to different tone guidelines.

@wawagilewski An AI marketing agency that launches branded campaigns is ambitious! The automation of campaign creation saves agencies and marketers massive amounts of time.

How does Protaigé maintain brand consistency across different campaign types? Are users able to input brand guidelines and tone of voice so the AI stays on brand?

Curious about how it handles campaign performance tracking and optimization over time.

digital marketers need to pay very close attention to products like this. Would love to give it a try for my side projects.

The connected workflow approach could really help avoid the disjointed outputs we see with multiple AI tools.

Excellent launch, Protaigé team. From a clarity & onboarding lens: when a marketing lead opens Protaigé for the first time, what’s the one belief you want them to hold in the first 10-15 seconds?
Is it:
• “I can launch a full on-brand campaign in minutes, not weeks.”
Or:
• “This system already knows my brand’s DNA—my work aligns automatically.”
Because in tools promising “speed + brand alignment,” the biggest adoption barrier isn’t features—it’s the user believing it fits me. Curious how you’re framing that for first-time users.

do you use a variety of different models? or one in particular?

Congrats on the launch Wawa, Protaigé looks strong and well designed. Upvoted. I am co founder of bestofweb .site and after your launch if you want more visibility you are welcome to join and introduce it to our founder community.

nerd alert: @orcdev turned @shadcn/ui into a beautiful 8-bit styled component library.

for me, it's an instant crush.

the project is backed by @Vercel's OSS program, sponsored by @Shadcnblocks, @Trigger.dev, and more; free and open-source. lfg!

Love it

I don't have a need for it (now), but I love it :)

It’s a fun project, and genuinely useful if you’re into retro-style UI. @orcdev has been doing amazing work creating videos and content for the shadcn ecosystem and I highly recommend you follow him on Youtube, my bet is hes the next big creator and Devrel at @Vercel - Keep up the great work mate!

This is awesome stuff, Orc friend! Here to support your launch. Good luck! 🙌

Upvoted my man :)

Love the retro design, also that it's Open Source. Knew immediately it has to do something with shadcn 🫡

Love this! 8bitcn/ui brings a super fun 8-bit vibe to modern components. Clean, nostalgic, and really well executed. Huge props to the Orc — excited to see where this goes! 👾🚀

Great components library for people who build retro games or websites! Awesome work @orcdev 👏

This looks fun and my thought is it would be cool if there was a small library of prebuilt 8-bit themes because I often struggle to match colors when I try to make playful UIs.

Lets gooo big Orc :p

Let's go!!

Looks awesome, definitely want to try :)

Cool stuff

Really cool. Been using gemini all summer. Really fast and nimble. Great for day to day work. Only using gpt5 and Sonnet for harder problems. So this is a welcomed model to compete with the elite LLMs. I hope its included in the startup perk package 😅🙏

I am wondering how it handles ambiguous problems because my logic tasks often have multiple valid interpretations and I want to know if it can explore all possibilities without oversimplifying.

Really exciting to see this launch because my past encounters with AI reasoning felt limited and this model gives me hope that complex problem solving might finally be more intuitive and human-like.

this feels built for people who enjoy complex thinking. The whole design pushes beyond typical assistant-style replies.

Tried a few logic and it handled them with such calm accuracy. Felt almost like talking to a real problem solver.

So not yet for Pro users? I would honestly feel betrayed if I were paying for Pro and would like to test this new feature

The human-like reasoning hits especially well on science problems. Got super clean explanations with zero fluff. 🔥

Nice launch! Deep Think sounds like the “let’s solve the impossible” mode. What’s been the most memorable problem users asked it to solve so far, and how did they react when it actually delivered?

Hey ProductHunt! 👋

I’m Berit, co-founder of Korl. Thanks for checking out our launch!

This Slack agent actually started as a hacky internal workflow we built for ourselves. We were tired of trying to remember to log Jira issues after every customer call. We were building things our customers had asked for… but forgetting to close the loop when they shipped. We knew AI could automate this, and we didn’t want yet another system to log into.

So we stitched together a rough agent that reviewed our Fathom call transcripts, flagged feature requests, and pushed them into Slack for quick triage.

Then one of our customers saw it and asked, “Can we have that?” That’s when we realized this should be part of our product, not just an internal tool.

Today’s launch is that productized version. It:

• Captures feature requests automatically from call recordings

• Routes them to Slack so CSMs or Sales can add or update Jira issues from Slack

• Tracks progress and drafts personalized updates when features ship

We’d love to hear from you. What’s your current process for tracking customer requests? And what would make this agent more useful in your workflow?

Thanks again for your support and for being part of the Korl community!

Beautiful launch Berit. Korl feels simple and powerful. Congrats to you and the team.

Cool. And finally, do you allow the presentations to be exported to different formats?

Oof, this hits my “where’d that request go?” panic before QBRs. Pulling asks from Gong → Slack/Jira then nudging when it ships… nice. Also into the auto-deck angle. Curious how well it de-dupes and maps to existing tickets.

Impressive launch, Korl team. From a clarity & onboarding lens: when a customer-facing team opens Korl for the first time, what’s the one belief you want them to hold in the first 10-15 seconds?
Is it:
• “I understand each customer’s unique value path, not just their usage data.”
Or:
• “My presentation will reflect their brand, context, and issues—no generic slides.”
Because in tools aimed at personalization at scale, the biggest adoption barrier isn’t features—it’s belief that it gets the customer, not just the data.

Korl hits a major pain point for CS/AM teams: the grind of turning scattered product + customer data into polished, personalized decks and renewal materials. Rather than wrestling with spreadsheets, Jira tickets, Slack threads, and having to build each customer‑specific slide by hand, Korl pulls everything together and auto‑generates meaningful presentations.

I especially like that it doesn’t treat “presentations” as generic templates — it builds them around real context: who the customer is, how they use your product, what their priorities are, and what value you’ve delivered or could deliver next. That shift from generic to personalized is where automation actually adds value.

For startups or small SaaS companies that can’t justify a full‑time CS ops or presentation builder, Korl seems like a tool that lets you punch above your weight: better customer communications, stronger renewals, and more consistent value messaging without scaling headcount.

That said — the real test will be how well the “AI + data sync → presentation” pipeline handles edge cases, complex data, and constantly evolving products. If that holds up, I think Korl could be a game‑changer for customer-facing teams.

I could see how our tool could benefit from Korl’s workflow automation and streamlined collaboration features, so I’m thinking this could work well together and I’ll go take a closer look at their launch.

How customizable is the routing to Slack channels or Jira projects?

@berit_hoffmann Tracking and fulfilling customer requests directly in Slack is perfect for customer success teams! Keeping everything in one place reduces friction and speeds up response times.

How does the agent prioritize feature requests when multiple customers ask for similar things? Are teams able to connect this to product roadmap tools for seamless workflows?

The AI component could be huge for identifying patterns across requests.

This is definitely useful. A lot of customer complaints are about not getting quick responses and no one taking charge.

Bun is around for quite some time, I wonder why it never made it to PH. Also, haven't they been bought by Anthropic now?

I ❤️ bun. But I made sure my project also works with NPM. Turned out. Bun didn't support windows arm executables. But NPM did. So for win arm exe. I use NPM. For other targets. Bun.

Impressive launch, Bun team. From a clarity & onboarding lens: when a developer opens Bun 4 for the first time, what’s the one belief you want them to hold in the first 10-15 seconds?
Is it:
• “This runtime just works with my existing stack, no rewrites.”
Or:
• “I can build faster and ship earlier, and the tooling problem is solved.”
Because for developer tools, the adoption barrier often isn’t features-it’s the dev believing this fits how I already build. Curious how you’re framing that.

Seeing a tool combines, task running, and npm support makes me smile because my old workflows were a patchwork of half a dozen utilities.

It would be cool if the runtime gave me quick feedback on errors because I hate waiting for the build to finish just to find a tiny typo.

Congrats! Bundling, running, and installing in one runtime is an elegant developer DX win.

Your landing page is partly broken

pricing feels fair, especially for small projects that still need professional delivery.

Do you offer tracking, and also is there a way to detect if the mail has landed in spam?
Would love to try it

So the only thing this one better than Resend is the number of free emails?

Very cool product :)
Looking forward to trying it!

Nice project, @bittucreator! It’s much-needed. I also love the simple and clean UI. Congratulations on a successful launch!

Do you have any plans for a PHP/Laravel package or library that can be used to send emails using UnoSend?

Well done Venkat. Unosend is clean and practical. Congrats to you and the team.

@bittucreator @Unosend 99.9% deliverability for transactional and marketing emails is impressive! Email deliverability is such a pain point for businesses, especially as inbox providers get stricter.

How does Unosend achieve such high deliverability rates? Are there built in tools for domain warmup and reputation management?

Curious about how it handles bounce rates and spam complaints to maintain that deliverability level over time.

landing page looks pretty wack on darkmode firefox.

as a paying resend user, why should i switch to unosend?

How does deliverability compare to Resend? Are you likely to end up in spam for gmail or outlook? Additionally, do you have any protection in place to prevent people abusing the service, and then get the other people using the service to be marked as spam?

@bittucreator I will definitely give unosend a try!

My recommendation would be to change up the UI a little, it feels like a clone of resend (which I currently use) and I think that could hurt the trust of customers who already know Resend.

@bittucreator @Unosend Looking good and feature packed, pricing also makes sense. Two questions though -

1. Is there plan for drag and drop templates?
2. Why is regional pricing only for India, when there are other countries where PPP is needed even more?

I signup and add domain but email are not going to recipient

on api response showing success

{"id":"ba75eee6-853e-4b60-8cb6-a0834a9b0a47","from":"verification@no-replay.mingleq.com","to":["sankalpchoudhary8@gmail.com"],"created_at":"2025-12-05T06:04:33.245466+00:00"}

Show raw log

Running into this issue when setting up outgoing mail with Supabase

In your case, Supabase’s auth service is doing this:

1. Connects to smtp.unosend.co on whatever port you specified.

2. Tries to use TLS (either because you told it “SSL/TLS” in the settings or because it does STARTTLS if the server advertises it).

3. During the TLS handshake it checks the certificate.

4. The server presents a cert whose hostname is localhost.

5. Supabase compares that to the host you configured smtp.unosend.co and goes “nope” and throws:

tls: failed to verify certificate: x509: certificate is valid for localhost, not smtp.unosend.co

I don't see a way to contact support anywhere on the site? So I am posting here to hopefully make the devs aware of this issue and resolve it?

Really curious to try this out, the reasoning improvements sound solid.

The Golden Kitty nod shows impact. What’s next, maybe expanding search into immersive AR or VR contexts?

Excited to try this out!

Hey Product Hunt! 👋 I'm Adit, founder of TypMo. In the age of AI coding tools, I noticed a problem: we're spending significant time and money prompting without structural clarity. The wireframing stage is where experimentation should happen, it's fast, cheap, and low-risk. So I built TypMo—where wireframes become implementation prompts.
It's Markdown for wireframes, a simple text syntax using design vocabulary you already know. No learning curve. Just type and see your UI render instantly.

With TypMo you can:

• Generate wireframes from prompts, sketches, or PRDs

• Run IA/UX audits on your structure

• Iterate on hierarchy, flows, and components freely

• Export implementation prompts when you're ready
  

That's where TypMo lives - the cheap zone where messy ideas become organized, prompt-ready structure. Experiment freely, validate fast, then let AI build exactly what you need, the first time.

Clarity before code. Wireframe first, prompt with precision.

I would love to have your feedback! Thank you!!

Super nice. Congratulations on the launch!

Sounds really interesting! How is the business model behind?

Translating wireframes into UI prompts – I really like the thinking and agree with the problem this is trying to solve. I will definitely give it a try!

Congratulation @aditgupta Love the “Markdown for wireframes” idea - that’s a super clean mental model.
How well does TypMo handle complex multi-screen flows?

Very unique and convenient. Check if I understand it correctly, we need to write our structure and it designs a wireframe and the output is the prompt that we can feed into cursor?

Congrats on the launch!

Great idea with the year-end summary. Can the app also suggest a plan, motivation, and goals for next year?

I've tried every journaling app out there—Day One, Journey, you name it. They all felt like digital storage boxes. Reflection feels like a conversation. The AI prompts aren't generic; they actually pick up on what I wrote yesterday and ask follow-up questions that make me dig deeper. It's like having a therapist and a life coach in my pocket. The cross-platform sync is flawless—I start on my Mac at work and finish on my phone in bed. Truly a game-changer for my mental clarity.

I love the concept and great timing on bringing it on here one day after the wrapped came out of Spotify 🤣, but i do believe that for us to understand what's to come, it's very important to revisit the past, and i love how you're product helps the users do a context dive. All love, wish you the best.

I am curious tho, what parameters are you reading the delta for?

Congrats on the launch. Many journaling apps struggle because reflection still feels like a task. From an onboarding & retention perspective, what’s the moment when a user actually feels the reset, not just sees the prompts? That feeling is usually what drives long-term use.

Hey Product Hunt! I'm Leila, the founder of Stardrift.

Have you ever put off booking a trip too long because it's such a pain - and paid for it by buying last-minute tickets at a steep price?

It’s complicated! You need to check your schedule, weigh off different price/time comparisons, and think about every leg of your trip and make sure your connections work.

In an age of AI, it's crazy that there isn't a better solution. That's why we built Stardrift, your personalized AI travel planner. Stardrift can:

• ✈️ Pull live hotel, flight and Amtrak prices & availability

• 💵 Help you construct an end-to-end itinerary, pricing out and comparing each leg

• 🗓️ Connect to your calendar and find the details of any meetings or conflicts

Even better, it's entirely free and available today - no waitlist or demos. Sign up and try it out today at stardrift.ai!

Great launch Leila. Stardrift feels sharp and useful. Congrats to you and the team.

I always put off booking then pay the panic tax. If it really pulls live fares and checks my Google Cal, I’m in. Curious about multi-city and tight Amtrak connections. Trying it on my NY-DC-Philly run next week.

It was always a headache opening 10 tabs to compare flights and hotels. Thanks for building this

This is clutch! Love that it syncs with your calendar.

The “book” button just redirects to an airline website, not to a specific flight. Is that by design?

I’m excited to be a power user - congrats on the launch; this looks awesome!

This is amazing @leilaclark ! Will we be able to also get restaurant/event recommendations?

Great launch, Stardrift team. From a clarity & onboarding lens: when a frequent traveller opens Stardrift for the first time, what’s the one belief you want them to hold in the first 10-15 seconds?
Is it:
• “This trip will work optimally for me, not just a flight booked.”
Or:
• “This assistant already knows how my schedule and constraints matter.”
Because in complex-trip tools, adoption hinges more on mental alignment than features. Curious how you’re shaping that.

Congrats! Besides transport and hotels, are you planning to add full trip itinerary planning too?

Hi everyone!

@Google Nano Banana Pro has just met a serious challenger. Seedream 4.0 was already impressive, but 4.5 takes multi-image fusion and dense text rendering to a completely new level.

It nails the details when blending multiple inputs. The typography is crisp and it finally handles complex layouts without messing up the text. Thanks to the "Spatial Logic".

The industry consensus is clear. The next phase of image generation is about logic and reasoning instead of just pixels. Top-tier multimodal models lower the barrier for visual creation. Seedream 4.5 makes professional expression efficient and accessible for everyone.

@Higgsfield , Fal, and @ImagineArt already support 4.5. The list is growing!

Just tested it and it's realy good. I still think google Nano Banana Pro is better with more realistic images and multiple people in an image. But price-wise Seedream 4.5 beats Banana Pro and they are trying to really close the gap to google's SOTA image model.

I think the images are fire, I love this tool. And what you say is actually true Nano banana Pro even though is better I think you two are competing.

Any chance you’ll be adding more languages in the future?

The realism of the photos is honestly amazing••••

Very realistic gallery. Really impressive

It is so good that it is scary. Crazy how AI can replicate visuals to such an extent.

Congrats on the launch! Claude Mem looks like the next step in making AI actually meaningful over time. What’s the moment users say, “Oh wow—it remembered what I said last week”?

Wow! This is really amazing idea.
So does it stay inside an IDE as an observer or inside a Git repo?

I usually store research conversations as markdown. And give it to an AI to then implement in the code. Is this what this sort of does, but more automatically? Or how doe sit work?

You nailed minimalistic design!

Hi Product Hunt!

We first launched Weather mini for Mac here in 2020. Since then, a ton has changed — and today we’re back with Weather mini 3, a lightweight weather app for iPhone, iPad, Mac, and Vision Pro.

What’s New in Version 3

• All-new weather illustrations with animations

• New widgets & Dock icons

• Trip forecasts with Apple Intelligence

Trip forecasts is a new feature for iOS and iPadOS. (coming very soon to Mac and Vision Pro)

It lets you check the weather for places and dates all at once, so there’s no need to jump between apps when planning a trip, especially to unfamiliar places.

With on-device Apple Intelligence, you can just type something like “Tokyo and Hong Kong next week” and get trip insights like daily weather, summary, notes, city context, and clothing tips in a few seconds. All happens right on the device since we just use system frameworks like WeatherKit, no 3rd party services at all, which means: there’s no huge LLM downloading, no data sharing or uploading, no usage limits in the app.

That’s it, new design, same lightweight weather app, now available on more Apple devices.

Please share your thoughts and feedback with us in the comments if you try it.

Such beautiful illustrations. Just love how the color palettes and tones combined with UI.

Will try out your app before planning my skiing trip. Thanks!

Wonderful. Something I was looking for a long time.
Will try out

Oh wow I love the UI design here. Really amazing! Great work guys!

Why an illustrated weather app?

As the designer behind Weather Mini, I wanted weather to feel less cold and data-heavy.

Even on a rainy day, a little illustration can brighten the moment.

If Weather Mini can give someone a tiny lift when the sky looks gloomy, that’s the goal.

When designing our AI features, we avoided third-party cloud services and heavy local LLM setups. Instead, we rely on what’s already built into users devices — no accounts, no downloads, no configuration. The goal is simple: a truly easy, “just works” experience that’s good enough to be useful in real life.

Great to see Weather mini for Mac back and stronger. From a clarity & onboarding lens: when a user opens the app for the first time, what’s the one belief you want them to hold in the first 10–15 seconds?
Is it:
• “I can glance at this and immediately know what matters for my day.”
Or
• “This isn’t just weather—it fits my workflow, right in the dock.”
Because for glanceable Mac utilities, the difference between “nice app” and “essential app” often comes down to that first felt alignment.

Thanks for checking out Weather mini 3!

I’m Kai, developer of Weather mini, here with @Ann, who’s the designer of Weather mini. We are happy to answer your questions about development, new design, on-device AI, etc.

Congrats on the launch guys! I’m so excited to use this 🫶

Having an agent that actually assembles everything together saves hours. love how smooth that feels 😌

Are we able to host the games we build for others to play? I couldn't catch that info. from the homepage.

This is incredible and super bullish on the team!! Congrats on the launch, be right back, going to make some games now

I tried the application and played and created a simple game. It was able to fix the bugs in the game after a few prompts.

Let users export the game and publish it as HTML or something. You can monetize that like Bolt or v0

Solid launch, Orca 3 team. From a clarity & onboarding lens: when a developer opens Orca 3 for the first time, what’s the one belief you want them to hold in the first 10-15 seconds?
Is it:
• “This tool actually aligns with how I think and build, not just how someone expects me to build.”
Or:
• “I can trust this model/tool to scale with my workflow, not hold me back.”
Because in high-performance dev/AI tools, adoption hinges less on specs and more on the belief that it fits me. Curious how you’re shaping that moment.

I can’t click on any of the games or assets shown in the home page.

Can't wait to try it. The website won't let me login

Great launch, IMAI Studio team. From a clarity & onboarding perspective, when a creator opens IMAI Studio for the first time, what’s the one belief you want them to leave with in the first 10–15 seconds?
Is it:
• “I can turn my idea into a visual prototype in minutes.”
or
• “Design isn’t the barrier anymore—I’m in creative control.”
Because in ideation-to-prototype tools, the adoption lever isn’t just features—it’s the user believing they can visualize their idea. Curious how you’re shaping that belief.

It feels like an absolute cost saver for e-commerce products. Great job
Any other category are you targetting?

I honestly didn't know H&M and Tata use IMAI! That's a really positive product validation!

I am sure you must have identified innumerable use cases for IMAI Studio in retail/e-commerce space, @viola_schritter

What has been you experience with Asian retail space w.r.t the usage of this tool? I mean, were you able to penetrate into the global markets so far?

Great work, I tested out IMAI for a while, and I think there's lots of potential use cases (product ideation, product design, marketing). What are the major next features on the roadmap @viola_schritter ?

Congrats on the launch. Quick question, is IMAI being used more for early concept validation, or more for instantly generating final marketing assets at scale?

https://www.youtube.com/watch?v=1Jb8OwDJOIU

Hi everyone, here’s a quick walkthrough of IMAI in action - how we go from a single product photo to multiple design ideas, studio shots, and ad-ready visuals in minutes. Would love your thoughts on what feels most useful (and what we should add next). 🚀

Planning the composition and visualizing the final result usually takes so much effort. This seems like it would be really helpful for that part of the process.

Congratulations on the launch! After trying IMAI Studio, I'm convinced this is the future of product creation.

Huge congrats on the launch! IMAI Studio is a game changer. The future of product visuals is here.

Hi everyone!

For a long time, the drone market has felt a bit stagnant. We’ve been choosing between "flying tripods" that require careful framing or FPV drones that require months of practice. Antigravity A1 breaks that dichotomy.

Incubated by @Insta 360 , this is essentially an 8K 360° camera that flies.

The Invisible Drone tech is magic. Because of the lens placement, the drone stitches itself out of the footage completely, leaving a pure, unobstructed view of the sky.

You can fly first, then frame later. You don't need to stress about composition mid-flight. Just put on the Vision Goggles, use the motion controller to point where you want to go (literally just waving your hand), and enjoy the view. Just pick the perfect angle—forward, backward, or top-down after you land.

It brings the wonder back to flying. It’s less about being a cameraman in the sky and more about being a bird. Plus, at 249g, it’s regulation-friendly in many regions.

Yes, we finally get a drone that feels like play, not work :)

Amazing! How on earth you got this below 250g I will never know - kudos to the design and manufacturing teams. I'll be ordering one shortly 👍🏻

Could future updates add AI‑assisted framing suggestions to make editing even smoother?

Almost thought this would be a launch of Google's coding agent Antigravity 😅 But damn I realy want an FPV drone and feel like I'm flying myself through my neighbourhood but pricing is unfortunately too big for my budget. I hope that these type of drones could go down to like $300 and I would be so hyped. But still, this drone looks sick, also with the 8k 360 video ability