... doorbell. It rings the same whether it's a package or a fire. That's basically the gap Pushary fills, same hook idea you've already wired, but the event carries a real handoff (done vs blocked vs

... tolling overall. With my team we often do reviews of our favorite new products (from product hunt and from other places) internally so happy to check out yours! Linkedin: https://www.linkedin.com/in/niko... Nikolas Comment from Ely(@ely8): @sk_uxpin  I also ask Claude to generate both unit and integration tests, but I rely a lot on behavioral/UI tests as well especially ones built with Maestro. They ve become a really important safety net for end-to-end user flows ... ... where most testing problems actually begin. Comment from Nikolas Dimitroulakis(@nikolas_dimitroulakis): @elara_thorn  I totally agree. things are similar in the dev tooling space + whats key is test automation and scripting. Comment from Stan Kolotinskiy(@sk_uxpin

... your audit database. The second is dramatically more defensible but most governance tooling treats them identically. In LineageLens , every record gets a confidence score from 0.0 to 1.0, broken into five independent evidence signals. Easy Mode captures (VS Code extension, no proxy) score around 0.27 honest about what you know. Power Mode captures (proxy running, full request interception) score up to 1.0. The score is not about whether the record is useful. It is about how much you can defend ... ... also like the idea of keeping weak signals visible instead of hiding them. A 0.27-confidence record can still say look here, as long as the UI prevents it from pretending to be proof. Comment from Stan Kolotinskiy(@sk_uxpin

... next. If that context disappears, it becomes much harder to understand how a decision was made, even when the AI history is still available. Curious how other teams are thinking about this. Comment from Stan Kolotinskiy(@sk_uxpin): @grover___dev  to me, this looks like a matter of an incomplete process - any refund approval or other important business decision should be fully documented so that the info is available in 6, or 36, or whatever months from ... ... decision maker, it only suggests doing something, but then there's a human who makes the decision - and in the case the lack of documentation/information is not related at all to the agent Comment from Stan Kolotinskiy(@sk_uxpin

... wrong, you get a clean human sentence instead of a wall of raw 502 Bad Gateway HTML. Smaller things that add up Sponsor and release-notes links are now real OSC 8 hyperlinks clickable in iTerm2, WezTerm, Kitty, recent VS Code; plain text elsewhere. Paste into the filter works now, including multi-rune batches. Backspace is multibyte-safe, and search input is sanitized against ANSI/C0 escapes. The PR diff viewer honours the monochromatic theme contract (monochrome / phosphor / amber ... ... Upgrade: brew upgrade gfazioli/tap/octoscope https://github.com/gfazioli/octo... As always: read-only, free, MIT, on Linux / macOS / Windows. Curious how long do you leave terminal dashboards open? Always-on second monitor, or open-check-close? Comment from Stan Kolotinskiy(@sk_uxpin

... when we did a launch, we were hounded by people trying to get us to go with them. So I don't know really how you would stop that. Comment from Stan Kolotinskiy(@sk_uxpin): We need moar topic forums! Comment from Stan Kolotinskiy(@sk_uxpin

... kinds of services and the way they massively collect and use information about users? In the EU, especially, this has become quite a big topic. Which services have you decided to stop using, or replace with more privacy-friendly alternatives? Comment from Tina Chhabra(@tina_chhabra): I run marketing so I'm kind of on the other side of this. the amount of user data you can buy from B2B providers without any real consent is honestly uncomfortable. we stopped ... ... Gmail) :D Comment from Nika(@busmark_w_nika): @jim_jeffers  I think many people are complaining when some company collects data but most of them use the service anyway (social media) :) Comment from Nika(@busmark_w_nika): @sk_uxpin

... github.com/gfazioli/octoscope/releases/tag/v0.18.0 As always: read-only, free, MIT, on Linux / macOS / Windows. Source: https://github.com/gfazioli/octo... Curious what preset would you add to the work filters? "needs my review"? "no release in 6 months"? Comment from Stan Kolotinskiy(@sk_uxpin): As always - great job! Comment from Giovambattista Fazioli(@gfazioli): @sk_uxpin

... Tedious backlog stuff goes to Codex. Still haven't figured out when to stop letting the agent run and just write it myself. That one's still on me. What's your current stack? Comment from Stan Kolotinskiy(@sk_uxpin): Basically Claude Code in manual mode for everything :) Comment from Theron Elric(@elrictheron): Totally agree. The biggest unlock is realizing these tools aren't interchangeable. Routing by task instead of forcing one tool to do everything saves ... ... from Devin Mancini(@_devinmancini): Love the task-routing approach. I ve found that sticking to a smaller stack is better. Though, switching between 4+ tools often creates more friction than it saves. Comment from Nolan Vu(@nolan_vu): @sk_uxpin

... that logs prove something happened, but they do not always explain why it was trusted. That context around approval feels just as important as the event itself. Comment from Dev Grover(@grover___dev):  I see your point @sk_uxpin . In many cases, the final human approval is the thing that matters most. Where I think it gets a bit different with LLMs is the volume and distance from the original reasoning. A human may approve the decision ... ... knowing which model produced it. As AI becomes embedded in workflows, accountability needs to include decision history, not just technical history. The way behind approval is often where the most valuable context lives. Comment from Stan Kolotinskiy(@sk_uxpin

... this series were brought back, which big-name founder or entrepreneur would you most want to see here and ask questions to? What would you ask them as the 1st question? Comment from Nika(@busmark_w_nika): @sk_uxpin  To be honest, I do not know him, but gonna google him :D Comment from Gabe Perez(@gabe): @byalexai   @busmark_w_nika  I did an AMA on that very topic in the past! https://www.producthunt.com/p/pr... Comment from ... ... Stan Kolotinskiy(@sk_uxpin

... domain knowledge making you more powerful than peer. this is a bold statement lol also, can you share me the work of that HTML file? Do you know where to host it? Comment from Nathan Tran(@nathan_tran9): @sk_uxpin  yeah this is the part i find interesting btw what's the dirtiest one-off you've handed off lately? ( ;) Comment from Nathan Tran(@nathan_tran9): @briancassingena  ha, you said the magic words. same story ... ... daily work. A quick HTML report or a cleaner client proposal can be enough. It makes the idea easier to share and you don't have to wait for someone else to build it Comment from Stan Kolotinskiy(@sk_uxpin

... your own makes it easier to stay on top of both without context-switching to the GitHub UI. Curious do you work mostly on your own repos or contribute to team projects? Comment from Giovambattista Fazioli(@gfazioli): @sk_uxpin  Thanks Stan! That s a really useful data point 1.5h/day just on the triage-before-the-review is significant, and it confirms something I suspected: the bottleneck often isn t reading the diff, it s figuring out which diffs ... ... close to how watch_repos works today for the dashboard itself. Out of curiosity, is that manual list something you keep in a doc, or is it more of an informal team convention? Comment from Giovambattista Fazioli(@gfazioli): @sk_uxpin

... already be partially automated with AI. Blue-collar work is physically harder to replace at scale because robotics is expensive, slower to deploy, and depends heavily on real-world environments. But I don t think it becomes humans vs AI. More likely: small teams + AI replacing large teams without AI. Comment from Haritha Vijayakumar(@harithavijayakumar): @busmark_w_nika  Yes, that s another huge shift. AI increases productivity, but at the same time it also changes perceived value and pricing ... ... transition will see probably see job hikes in other areas before we see real lay offs in productivity sense Comment from Johnson Gill(@johnsongill): @busmark_w_nika  Uncertainty with a lot of questions. Comment from Stan Kolotinskiy(@sk_uxpin

... fire/hire? :D Comment from Nika(@busmark_w_nika): @thomas_gdpt  This will be the standard for each private business. And it seems that gov positions will be the stable ones. Comment from Nika(@busmark_w_nika): @sk_uxpin  The only security is created synthetically, in gov positions. Comment from Nika(@busmark_w_nika): @evitam  There will be IMO, fewer companies eating each other, with fewer employees trying to outperform any other company that uses AI. Very strange ... ... time. Comment from Nika(@busmark_w_nika): @lisa_steingold2  Just a few minutes ago read this post: https://x.com/kittenheelsrule/st... . The situation is so desperate. Comment from Nika(@busmark_w_nika): @sk_uxpin

... launching on PH soon... would appreciate a follow (See PRODUCT HUNT LAUNCH link in my profile). I'm building The Sponge, an AI powered flashcard app that turns any webpage into study material. Comment from Stan Kolotinskiy(@sk_uxpin): Good luck, that's indeed impressive! Comment from Tudor Pop(@tudorp06): Congrats, for 16 years old a project in that sector is a big thing! Hope it works out and you get at least a few first users ... ... glad to know more about the app. Comment from Matthew Goley(@matthewgoley): @sk_uxpin

... Following up on what @sk_uxpin asked for on the v0.11.0 launch thread: you can now read pull-request diffs inside octoscope. Press Enter on any PR for the drill-in. Press f for a full-screen list of the changed files. Enter on a file row opens its unified diff with syntax highlighting chroma's diff lexer, monokai palette, the universally recognised red/green for - / + lines and a scrollable viewport so a 400-line patch fits without pushing anything ... ... github.com/gfazioli/octoscope Curious what people drill into next. Side-by-side diff view? Ignore-whitespace toggle? Per-file folding? Let me know the v0.11.0 thread shaped this release, the v0.12.0 thread will shape the next one. Thanks again @sk_uxpin

... simincident.com . No programming background apart from casually learning C++ years ago at a basic level. The AI wave was a game changer for me because I always wanted to build my own product. Started by pasting ChatGPT code into VS Code, then discovered Claude in February this year and that was it. I was able to deliver a whole product. The problem? I never validated if anyone actually needed it. Zero traffic. People who signed up for beta never tried ... ... when I started building Luminens a 16-bit browser RAW photo editor running on WebGPU and libraw-wasm. Talk about a nightmare of structural complexity. I moved away from automated builders (I've tried all of them), switched to VS

... that looks great in a demo and then it falls apart the second real users start hitting edge cases. the starting is easy part. keeping it together at month 3 is where everything breaks Comment from Stan Kolotinskiy(@sk_uxpin): Well, code quality in itself is still really far from ideal (not to mention the consistency indeed) Comment from Haritha Vijayakumar(@harithavijayakumar): @sk_uxpin

... experiment for my startup, @Pretty Prompt : Remove the monthly tier; offer only Free or Yearly (basically no monthly plan available). We decided to do this after doing some simple maths and seeing that 42% of users today choose annual vs our monthly plan. So what if I can take out the monthly option entirely? Has anyone here done something similar or has experience with this? I think a big part of PMF is also a different type ... ... Pricing Market Fit). Does removing monthly pricing hurt conversion or actually make pricing clearer and push more users toward longer commitment? So then I did a bit of research: @Screen Studio Monthly + heavily discounted annual ($29/mo vs

... Yesterday s discussion about who is more likely to be replaced (white-collar vs. blue-collar workers) raised another question for me. So I m asking: What do we need to do to keep our jobs in the age of AI and robotisation? What skills do we need to master and become self-sufficient in and better at than any machine? Just for fun, I thought about how students at the University of Arizona booed Eric Schmidt after he told ... ... role. I think the best way in this era is to identify your niche, train enough and start working on building your own brand, because if you are an entrepreneur then AI is your friend, not a competitor

... understanding. Curious if others have felt this too. Did AI coding assistants ever make you feel disconnected from your own product? Or did you find a healthier balance that still keeps you in control? Comment from Stan Kolotinskiy(@sk_uxpin): @muhammadabdullahkhan  yeah, definitely a fair point, and I'm pretty sure that there is more to come, but for now IMV it's definitely better to stay focused and mentor the LLMs rather then letting them go wild Comment ... ... from Muhammad Abdullah Khan(@muhammadabdullahkhan): @sk_uxpin

... hard, as I can do everything faster. But maybe I just felt bad:))) At least, it's really possible to work 2-3 times more and earn too, if organize everything right. Comment from Mona Mehta(@mona_mehta): @sk_uxpin  I would tag this as a benefit, especially the reading part to refuel your batteries :) Comment from Nika(@busmark_w_nika): @gabriel_brooks1  So basically, it can redefine priorities. Comment from Alper Tayfur(@alpertayfurr): I feel this ... ... work expands to fill the time you freed up. However, AI is not something that always guarantee quality and that's why the human role for auditing and checking is still important for me Comment from Stan Kolotinskiy(@sk_uxpin

... might be selling the technology instead of the transformation. Would love to hear from other founders: have you experimented with de-emphasizing AI in your positioning? What happened? Comment from Mona Truong(@monatruong_murror): @sk_uxpin   Totally get that. It's become background noise at this point. The irony is that the more everyone slaps "AI" on everything, the less it actually means to people. We found that just describing what the product does without the label made users ... ... about how it works. They care about how it makes them feel. Glad the distinction resonated with you. Comment from Mona Truong(@monatruong_murror): @edikan_peters   Love the fintech example that really nails it. "Real-time transaction monitoring" vs

... should train AI and then be its overlord. Comment from Zakaria empire(@zakaria_empire): I m building Le morceau de prodige with React/Node right now, so this hits close to home. For me the shift isn t AI vs me , it s me + AI vs