2025年10月2日 の Show HN

I’ve been working with @artf (creator of GrapesJS) on Grapes Studio, an HTML-first editor with an LLM assistant on top of GrapesJS.

We’re approaching this differently than the new wave of AI app/site builders which are typically generating full React applications, which we think is overkill for simple websites. From talking to people using these tools, we’ve seen a lot of issues with build errors and overly complicated pages.

With our approach you can:

- Edit visually via the no-code editor (drag/drop) or ask the LLM to make scoped changes (like “add a section” or “add a new page”).

- Build with straight HTML/CSS

- Ask AI to import your current site and start building from there instead of total rebuild.

We think there’s a lot of benefit using drag and drop editor functionality with LLMs, or you can jump straight into the code in the editor if you choose.

- Do you see value in this hybrid model (AI + visual + code editing)?

- What are the biggest blockers you’ve run into with AI-only builders?

Let us know what you think.

Hey HN! I built YNOT, a simple cross-platform YouTube downloader with a GUI.

It's powered by yt-dlp and completely free/open-source (WTFPL license).

Key features: - Simple GUI - just paste URL and download - Downloads HD/4K videos - Cross-platform (Windows, macOS, Linux) - No ads, no tracking, completely private - Lightweight and fast

GitHub: https://github.com/james-see/ynot

I'd love to hear your feedback and suggestions!

My friend and I built BookMyFlight to finally modernize flight search + booking.

Why we built this:

- Personalization. I fly the same route every month, and there’s no platform that knows my preferences so that I can open it, find and book my flight, and close it within a minute.

- Booking is slow. I hate seeing a long clunky airline form each time I need to book. I want booking a flight to feel more like booking an Uber.

How it works:

1. Optionally make an account and save your traveler preferences. Personally, I've specified my routine route as SFO to CLE and that I only want red-eye direct flights for this route.

2. Search for flights using chat or the search panel. Chat feels especially time-saving when you have preferences saved (e.g. I just say “search my routine trip").

3. Once you find the flight you want, use the one-click book feature which books your flight directly with the airline. The first time you book a flight, you’ll have to fill out your traveler info, but you won't see that form after that.

Notes:

- Your booking is directly with the airline (this means when something goes wrong, you get direct support from the airline—not a third-party)

- You can add your rewards numbers for each airline to keep earning points/status

The ultimate goal is to create the best possible experience that every traveler wants, but that OTAs and airlines don’t care to create. Also very receptive to hearing pain points from frequent flyers; we think this space is really outdated and could use some innovation.

Try it out and let us know what you think :)

I'm very excited to share what I've been working on lately!

Introducing ENHANCE, a terminal UI for GitHub Actions that lets you easily see and interact with your PRs checks.

It's available under a sponsorware model. Get more info on the site:

-> https://gh-dash.dev/enhance

This is an attempt to make my OSS development something sustainable. Happy to hear feedback about the model as well as the tool! Cheers!

I built Silobase , an open-source backend-as-a-service packaged as an npm module.

With just a package.json and a .env file, you can deploy a REST API on top of your own database.

We saw that the existing solutions have a learning curve and restrictions. So we thought to assemble one that can make things easy. Let me know your thoughts on this.

Hi, everyone!

I built a TimeLock NPM Registry to prevent supply chain attack. I was inspired by minimumReleaseAge of the pnpm, but I'm using bun for my projects.

TimeLock NPM Registry is an alternative npm package registry focused on supply chain security.

Its core feature is introducing a time lock before new package versions become available for installation. This protects developers from compromised releases: while packages are “on hold,” the community and security tools have time to detect and block malicious code.

Why it matters Reduces the risk of installing malicious packages. Lets you “wait out” 24 hours or more before updating. Increases trust in dependencies and builds.

How it works

A package author publishes a new version. TimeLock NPM Registry places it into a pending state for a set duration (e.g., 24 hours). Only after the timer expires does the package become available for installation.

Tech stack — Cloudflare Workers, Honojs.

I built a small browser game in the style of memory/pexeso using images of Ubiquiti products. It's fan-made and non-commercial (I used to work at Ubiquiti) and made for my son to have fun. All product names and logos are trademarks of their respective owners. Would love feedback on UI/UX and performance tweaks.

Like many of us, I struggle with staying focused and avoiding distractions while browsing the web. I’ve tried all the website blockers out there – Blocksite, Freedom, you name it. Most of these tools failed for one of two reasons: they were too strict, so I ended up disabling them permanently, or they were too lenient, so I would quickly find a way around them.

To solve this for myself, I built DecayBlock, a browser extension based on the principle of adaptive friction. The mechanism is simple:

1. When you visit a site on your distraction list, a brief timeout is applied before the page loads.

2. This timeout accumulates with each visit, so repeated or habitual access becomes progressively more difficult.

3. Over time, the accumulated timeout decays with a configurable half-life when you stay away from those sites.

The goal isn't to enforce a ban (although you can permanently block websites if you choose), but to introduce a small, increasing cost to the act of procrastination. That brief pause is often enough to break the subconscious loop of opening a distracting tab, without ever fully locking you out. You can configure the timeout growth rate and decay half-life to match your own habits.

I've been using this system daily for several months, and it has been significantly more effective for me than any other tool.

DecayBlock is available now on both the Chrome and Firefox web extension stores. Currently it’s only on desktop but I’m hoping to release a mobile version at some point in the future.

Chrome: https://chromewebstore.google.com/detail/decayblock/lpljcnal...

Firefox: https://addons.mozilla.org/en-GB/firefox/addon/decayblock/

I'm here to answer any questions and would love any feedback!

Hi HN!

I built Bookmark GPT Pro to solve my own problem: I had hundreds of bookmarks but Chrome's search only does keyword matching.

What it does: - Full-content search - searches titles, headings, descriptions, and full page text (not just titles) - AI chat interface to ask questions about your bookmark collection - Example: search "performance" and find "Making React Apps Faster" even if "performance" is only in the page content

Technical approach: - Weighted keyword search (title=50pts, headings=25pts, content=10pts) - AI chat powered by OpenAI or Ollama - Ollama support means you can run completely local/offline - No server-side storage - privacy-first design

Why this matters: Traditional folder organization breaks down at scale. Chrome only searches bookmark titles. Full-content search finds bookmarks based on what's actually on the page.

Privacy: With Ollama, everything runs on your machine. For cloud models, we only transmit encrypted queries - no bookmark storage.

Chrome Web Store: https://chromewebstore.google.com/detail/bookmark-gpt-pro/de...

Troubleshooting Guide: https://g-savitha.github.io/bookmarkgptpro-info/troubleshoot...

Built this over a few months of weekends. Would love HN's feedback on the approach and implementation!

Made a simple SVG converter with a friend. Under the hood, it uses vtrace for tracing and some image optimizations to make the trace cleaner. All feedback is welcome :)

Hi HN,

I built Valoris Research to solve a problem I kept running into: there’s a flood of brilliant AI research on arXiv every day, but very little of it makes the jump into real, venture-backable startup ideas.

Valoris Research is a daily feed that:

- Scans the latest arXiv AI papers

- Evaluates them for market potential, customer pain points, and venture readiness

- Publishes concise, human-readable venture ideas inspired by the most promising works

Think of it as a bridge between emerging academic research and practical innovation — built for founders, investors, and technologists looking for their next big idea.

I launched this as a side project a few weeks ago and have been shipping daily. I’d love feedback from the HN community — especially around:

- Usefulness of the venture idea format

- Gaps or improvements you’d want as a technical founder or investor

- Whether there are other research domains I should cover beyond AI

Happy to answer questions about how I evaluate research, pick ideas, or scale the pipeline.

— Sumit