🗣️ Post 1: Show HN: Same prompt tested across Replit, Bolt, v0, Lovable and Raq.com

As posted by: hawke  |  🔥 Points: 6

🔗 https://raq.com/real-world-test

💬 Summary

Hi HN,

I built Raq.com – a platform that uses Claude Code to build working internal tools directly in the browser.

Claude Code is great at self correcting when given the right tools.

I've found that the popular web-based AI coding tools look great in demos but fail on real API integrations or require a lot of error back and forth. They don't appear to do much research or self-correcting, likely to reduce spend. I wanted to see the current state of these tools, so I ran the same prompt on five platforms (Replit, Bolt, v0, Lovable, and Raq.com) to build a tool that requires 3 different APIs (Companies House, FinUK and OpenRouter) working together.

Four platforms produced broken prototypes or needed manual fixes. Raq.com delivered a complete working solution from a single prompt (that can be deployed to live with one click).

Full test with videos: https://raq.com/real-world-test

We're in early access (requires Claude Pro/Max for free usage) - we're looking non-coders who would like to build internal tools for their team.

Some technical info:

- Raq.com provisions isolated dev and prod Docker environments for each company (companyname.raq.com and companyname-dev.raq.com).

- The dev site includes a persistent terminal streamed to the browser, so the session continues even while tab is closed.

- CLAUDE.md file provides best practices, known pitfalls, and coding patterns for the Laravel + Filament stack.

- Self-Correction Loop: Claude can test and debug its own work. It has direct shell access to a custom script that bundles PHPUnit, syntax checks, and cache clearing. Plus a Playwright wrapper to check for errors and take screenshots.

- A single click runs a script that rsync's the dev workspace to the prod container, runs migrations, and clears caches.

🗣️ Post 2: Google profits even as its AI summaries reduce website ad link clicks

As posted by: rntn  |  🔥 Points: 4

🔗 https://www.theregister.com/2025/07/29/opinion_column_google_ai_ads/

💬 Summary

Opinion Alphabet, Google's parent company, is making money hand over fist. In its latest quarterly report, Google's revenue grew 14 percent year-over-year to $96.4 billion. While Google's cloud revenue, $13.6 billion, with 31 percent year-over-year growth, is growing fast, the bulk of its cash, $54.2 billion worth, still comes from advertising. What I wonder, though, is whether, even as its quarterly ad revenue hit a new all-time high, this can possibly continue? You see, Google's move to place AI-generated summaries, aka "AI Overviews," at the top of search results is wreaking havoc on websites. How bad is it? Bad. According to a new Pew Research Center study of 900 US Google users, when they saw an AI Overview, they were...

🗣️ Post 3: Show HN: Talk-to-design tool for total beginners. 10k users in 90 days

As posted by: hilongjw  |  🔥 Points: 3

🔗 https://seede.ai/

💬 Summary

Hey hacker news, I’m Evan. Last August I left dora AI because I realized that for true “newbies,” Canva and Figma are still too complex. My roommate wanted to make a grand-opening poster for a cat café and got stuck in Canva for 40 minutes—picking templates, tweaking fonts, hunting for images—then gave up.

So I built a tool where you design just by talking. Twelve weeks later, in March 2025, Seede AI went live with a single promise:

“Describe your idea like a tweet, and get a print-ready, professional design in under a minute.”

🗣️ Post 4: Show HN: I created an AI writing app that helps you write books

As posted by: playsong  |  🔥 Points: 3

🔗 https://bookswriter.xyz/

💬 Summary

TL;DR: Built BooksWriter - an AI tool that generates novels chapter by chapter while keeping YOU in control. Generated my first book in 2 hours. Looking for feedback from fellow writers.

Hey writers! I want to share something I've been working on that might interest you.

I've been frustrated with existing AI writing tools for ages. Take the current best tools for writing as an example- it looks promising at first, but the quality and consistency completely falls apart as your story progresses. By chapter 5-6, it becomes nearly impossible to maintain coherence. The AI loses track of your plot, characters start acting inconsistently, and you're drowning in clichés and tired tropes.

Writing directly with AI models is even worse. The longer your story gets, the more context issues you hit. You spend more time crafting prompts than actually writing, and when you ask for edits, the AI rewrites everything and destroys the parts that were working.

I built BooksWriter to solve these exact problems:

Give it your idea, and it generates your story chapter by chapter. But here's the key - it gives you 3 different directions for each chapter, so YOU choose where the story goes.

Unlike other tools, it maintains consistency and quality throughout your entire book. Chapter 20 is as good as chapter 1.

When you need changes, it doesn't rewrite everything. It finds the specific moments that need editing and improves just those parts without touching what's already working.

Upload your own writing samples and it adapts to write exactly in your style.

Generate in 19+ languages.

Don't want to read? Listen to your story instead. You can generate audio of your book too.

And finally: One-click publish to our platform at read.bookswriter.xyz

I generated my first complete book in just 2 hours while maintaining my writing style and creative control. The combination of automation + control is what makes this different from everything else out there.

The app is currently in closed beta. If you want to test it out:

Visit bookswriter.xyz Sign up Email support@bookswriter.xyz for credits, and we will give you access And start writing!

Check out what other users have already created at read.bookswriter.xyz

Looking for honest feedback from the community! What features would be most valuable to you? What problems do you face with current AI writing tools?

If you're a writer, I genuinely think you should try this. Happy to answer any questions in the comments!

P.S. - I'd love to read whatever you create and publish on the platform

🗣️ Post 5: Show HN: AI Flag Cleanup

As posted by: GarethX  |  🔥 Points: 3

🔗 https://news.ycombinator.com/item?id=44721731

💬 Summary

Flag cleanup is one of the most annoying things about feature flags. Many vendors have created little helpers. Usually they’re CLI tools that highlight mentions of feature keys in your code, or some also show you which flags haven’t been accessed in a while. But these only make things less painful, you’re still on the hook to do the work.

So at Bucket (https://bucket.co) that’s the part we wanted to solve with AI, removing the flag code itself.

- How it works:

The cleanup bot operates as a GitHub integration with Bucket. When it runs, it searches for usage of the Bucket SDK in the codebase and identifies where specific feature keys are used. LLMs are employed to refactor the code to remove the flag and eliminate codepaths that become unreachable.

So all you should need to do is review and approve the PR. Once merged, the flag itself is archived in Bucket.

- The results:

We’ve been running this in production ourselves for a few months alongside a few early-access partners. We’re seeing consistently good PRs that can be merged without changes.

However, this wasn’t always the case. Early on in development, the range in quality of solutions generated by the LLM was an issue. Most attempts were fine, but then it would add a comment somewhere unrelated, remove whitespace or change the code formatting (even if you gave it rules to follow). Many of the early generated solutions failed CI due to unused imports, incorrect indentation, lone decoration not always getting cleaned up, or unused exports failing to be removed.

You can get a surprisingly long way with an LLM and a quick prompt, but getting good results, consistently, is where the real work is.

Some of the issues we were able to solve through better prompting and using different models, while others needed custom solutions. We wrote up how we went about building it here: https://bucket.co/blog/building-ai-flag-cleanup

- Try the beta:

We’re opening the beta up to everyone, but limiting it to React codebases for now.

If you’d like to try it out: - Sign up at https://bucket.co - Enable it: https://app.bucket.co/envs/current/settings/org-integrations

Let us know how you get on.

🎯 Final Takeaways

These discussions reveal how developers think about emerging AI trends, tool usage, and practical innovation. Take inspiration from these community insights to level up your own development or prompt workflows.