2025年7月27日 的 Show HN

Hey HN,

I tend to use printf-style debugging as my primary troubleshooting method and only resort to gdb as a last resort.

While I like its ease of use printf debugging isn't without its annoyances, namely removing the print statements once you're done.

I used to use trace-level logging from proper logging libraries but adding trace calls in every corner quickly gets out of control and results in an overwhelming amount of output.

To scratch my own itch I created dlg - a minimal debugging library that disappears completely from production builds. Its API exposes just a single function, Printf [1].

dlg is optimized for performance in debug builds and, most importantly, when compiled without the dlg build tag, all calls are eliminated by the Go linker as if dlg was never imported.

For debug builds it adds optional stack trace generation configurable via environment variables or linker flags.

GitHub: https://github.com/vvvvv/dlg

Any feedback is much appreciated.

[1]: Actually two functions - there's also SetOutput.

Hey! This is something i have been working on. A tiny neural networking lib to learn how something like pytorch works, and to improve my own coding standards.

Aria is a modern, dynamic scripting language. It is meant to be a "sweet spot" language, easy to pick-up and enjoyable to use.

It comes with a familiar C-style syntax, and draws inspiration from a variety of languages. It has a small but usable standard library and strives to be a low-ceremony-get-stuff-done kind of language.

It is currently at version 0.9 and I would love feedback as I work towards getting it to 1.0.

I have realized that a lot of people nowadays self-host services and set up home labs with mini racks.

One major pain point I have come across personally is to quickly get health status from self-hosted services and machines, and have the ability to headlessly control my Raspberry Pi inside a mini rack.

So It got me thinking about building a built-in GUI that users can easily add to their Raspberry Pi nodes in their (mini or full) racks or elsewhere.

I have previously designed this GUI for an open source project I have been working on (called Ubo pod: github.com/ubopod) and decided to detach/decouple the GUI into its own standalone module for this use case.

The GUI allows headless control of your Raspberry Pi, monitoring of system resources, and application status.

I am designing a new PCB and enclosure as part of this re-design to allow for a new form factor that mounts on server racks.

I am recording my journey of re-designing this and I would love to get early feedback from users to better understand what they may need or require from such a solution, specially on the hardware side.

The software behind the GUI is quite mature (https://github.com/ubopod/ubo_app) and you can actually try it right now without the hardware inside the web browser as shown in the video:

https://www.youtube.com/watch?v=9Ob_HDO66_8

All PCB designs are available here:

https://github.com/ubopod/ubo-pcb

Hi folks,

About seven months ago, via HN, I got nerdsniped into a silly guitar transcription problem and made a bunch of really senseless code but what came out of it was what I thought at the time could potentially be pretty useful - a guitar fretboard mapper and fingering scoring algorithm.

So as of yesterday morning I've finally put those bits of code to "good" use, creating gtrsnipe to convert between MIDI files (.mid) and ASCII tab (as well as VexTab and ABC notation) and any combination/direction among the set of formats.

gtrsnipe tries to intelligently find the best neck and fingering positions using a note to fretboard mapper and a scoring algorithm that is unavoidably shaped by my subjective opinions and skills as a player but it does its best to avoid objectively impossible fingerings.

See the example tabs and usage in the README and please, try your own transcriptions from MIDI and if you love or hate the arrangement it gives you, I'd love to hear about it so I can further refine the scoring algorithm.

Thanks!

Hi HN!

I'm excited to share Flyde 1.0. A big update to the open-source visual programming tool I launched here in March of last year (https://news.ycombinator.com/item?id=39628285).

Since Flyde’s launch, there's been a huge rise in demand for visual builders, especially for AI-heavy workflows. Visual-programming shines with async and concurrency-heavy logic, which describes most LLM chains perfectly.

A few months ago, I tried to capitalize on this trend by launching a commercial version of Flyde called Flowcode (https://news.ycombinator.com/item?id=43830193). It didn't go well. I learned the hard way that Flyde’s strength wasn't just about flexibility or performance compared to tools like n8n. The real value was always how Flyde fits inside your existing codebase. The launch also helped me understand that there's still a big gap: no tool really covers the full lifecycle, from rapid prototyping to deep integration, evaluation, and iteration inside your own projects.

So, over the last few months, I worked hard to polish Flyde: - Cleaned up and simplified the nodes API - Made it possible to fork any node for maximum flexibility - Launched a new online playground for quick experimenting and sharing (https://www.flyde.dev/playground) - Created a new CLI tool to speed up development and setup - Fixed a ton of bugs - Simplified the UI/UX to make it smoother and less confusing

There’s still a lot of missing stuff. Better templates, docs, and nodes, but I think it’s finally stable and useful enough to give it another shot.

My plan is to first make sure that Flyde is usable and valuable as an OS project, and then try to provide additional value via “Flyde Studio” - a SaaS that will help non-engineers iterate on Flyde flows from a web-app. Changes become a PR in the host repo.

I'd really love some honest feedback and hear whether Flyde resonates with an existing pain/problem.

Check it out here: Playground: https://www.flyde.dev/playground

GitHub: https://github.com/flydelabs/flyde

Looking forward to hearing your thoughts! - Gabriel

Hey HN! When I started looking into LLMs and agents for software development and introducing them at work, I quickly realised that a person new to the topic faces a real barrage: - all the hype (AGI, engineers getting replaced by AI etc.) - conflicting opinions in virtually every discussion—for every person saying they’ve 10x-ed their productivity, there is a comment decrying LLMs as an utter failure - a lot of jargon (MoE, MCP, RAG, distillation, quantisation etc. etc.) - a profusion of models, IDEs/IDE extensions, CLI agents, other tools etc.

Sorting through all of this can be quite tricky for somebody coming in fresh! HN users have been discussing LLM tools for a long time of course, but many programmers I know still haven’t _really_ tried anything other than an LLM web UI and haven't been following the progress of tools much.

So my goal for this project was to provide a balanced overview of the topic, point people to substantive resources on eg. context management and productivity effects, and cover the concerns and risks as well (from prompt injection to shady training data sourcing). I hope it's useful!

Hey Everyone! Avi and I (in the current YC S25 Batch) noticed how fragmented crypto compliance is with rules changing constantly, vary by region, and force developers and legal teams into a scramble.

So we hacked together a prototype framework that generates a “guardrail” smart contract which:

- Hooks into your KYC provider - Enforces rules defined a simple compliance.yaml file - Automatically blocks users you can't legally serve

When regulations shift (new blocked country, updated KYC threshold, etc.), just update your YAML and redeploy with no smart contract rewrites. Audit prep also becomes a breeze. Our goal is to make staying compliant as easy as possible

How it works:

1. Declare once in compliance.yaml modules: token_sale:

start_date: "2025-02-01" end_date: "2025-03-01" max_cap_usd: 500000 kyc_threshold_usd: 1000

geographic_restrictions: blocklist: ["US", "CN", "IR"]

2. Compile with one command shor compile --blockchain ethereum --with-oracle

This generates:

- Guardrail.sol – Smart contract that reverts non-compliant transactions - policy.pdf – Lawyer-readable compliance documentation - audit.json – Manifest with rules, timestamp, and bytecode hash

3. Deploy & forget - Drop Guardrail.sol into your existing deployment flow - Smart contract enforces rules automatically on-chain - Run shor export-audit --format=zip for auditors

Why now?

- Regulations are getting tighter and full, active compliance is becoming more necessary

Try it in either SDK or CLI: https://github.com/ShorPay/shor-compliance

We'd love your feedback on:

- Would this be useful for your projects? - How are you handling compliance today? - Which jurisdictions or rules give you the most trouble - Missing jurisdictions (we only have US/EU/SG) - KYC providers you need (we support Sumsub) - Smart contract patterns we should add

Email: [email protected]

Contributing: https://github.com/ShorPay/shor-compliance/blob/main/CONTRIB...

Hosted own Version of Omegle

As I worked on my personal projects, I kept running into the same problem: There wasn’t a simple, local way to convert screen recordings into GIFs that look clean in a README.md.

Recordings don’t render inline on GitHub READMEs, and I didn’t want to rely on bulky video editors or upload my files to some sketchy ad-filled site. So, I wrote a small script.

I found myself using it so often that I turned it into a proper tool: minimal, local, and built for developers who care about good documentation.

I’m not some elite hacker — just a dev who wanted a better way to share demos. Thought others might find it useful too.

From an Airbus A400M to Belgian Chococolate Testing LLMs for specific context domains and building in public to keep the experiment running

I built a simple tool to help founders figure out who their actual competitors are! You know... the ones your potential customers already know and compare you to.

Just paste your domain, and we generate a focused list of 10 competitors with names, sites, and a quick positioning note for each.

Why I built it: I run a competitor monitoring tool (https://champsignal.com), and I realized that before people can monitor competitors… they first need to find them. This is harder than you think for people that have not been around for years aha

(It's free and doesn't require signup)

Would love feedback, especially if you don't think it's giving you the right competitors. Happy to improve the product.

Byte-Vision is a privacy-first document intelligence platform that transforms static documents into an interactive, searchable knowledge base. Built on Elasticsearch with RAG (Retrieval-Augmented Generation) capabilities, it offers document parsing, OCR processing, and conversational AI interfaces.

ClosedLinks introduces a very simple yet powerful model for secure, anonymous sharing through one-time access links that leave no traceable sender fingerprint. Unlike conventional tools that prioritize persistence and traceability, ClosedLinks is purpose-built for interactions where privacy, control, and non-attribution are essential.

If interested to try it out, check out - https://www.closedlinks.com

If you want to read more on why I built it and how it works, check out the white paper - https://www.closedlinks.com/white-paper/

Open to suggestions and collaborations.

Starting a new SaaS project with Phoenix is a genuinely fantastic experience, as anyone who's dived into it will tell you. The framework itself is robust, performant, and a joy to work with. However, as I've been building out my own SaaS ventures, I've noticed a recurring pattern: no matter how brilliant Phoenix is, you inevitably find yourself configuring the same fundamental aspects time and time again. Setting up authentication, user management, payment gateways, and all the essential plumbing feels like a familiar, if slightly tedious, ritual.

That's precisely why I developed the Phoenix SaaS Kit. It's designed to eliminate that repetitive setup, letting you hit the ground running with a solid foundation. The beauty of it is its modularity – you can cherry-pick the features you actually need for your specific project. So, while Phoenix gives you the incredible power to build amazing software, my kit offers a shortcut, freeing you from the boilerplate configuration and letting you focus on what truly differentiates your SaaS.

What's included: Phoenix LiveView with modern UI compon ents Comprehensive modular setup system User authentication system with registration/login oAuth with Google ready to use oAuth with Github ready to use Multi-tenancy with organizations and role-based access Payment processing integration (Stripe/LemonSqueezy/Polar) Blog system with admin interface via Backpex Error tracking and analytics integration AI/LLM integration capabilities via Langchain Rate limiting and security features Modern styling with Tailwind CSS and DaisyUI Design system admin page Background Job Processing Custom Haskoy font integration Legal pages (Privacy Policy, Terms of Service) Customisable, modular marketing components Development tooling and Claude AI integration Claude AI codereview Github Action Optimised Claude Code Sub-agents and commands

One time purchase for unlimited uses by a single developer

Hi HN,

I built a WordPress MCP Server that connects Claude Desktop to any WordPress site using the Model Context Protocol (MCP). It's designed for developers who want to work with WordPress content directly from their editor, assisted by AI.

You can interact with posts, plugins, and themes through Claude — for example:

Browse and edit content contextually Debug and develop plugins with AI help Automate repetitive WordPress tasks It’s fully open-source and self-hosted: https://github.com/docdyhr/mcp-wordpress

I’m looking for feedback from developers interested in AI-assisted workflows, WordPress tooling, and protocol-based app integration.

Happy to answer any questions — and open to contributors!

– Thomas

I made a little tool to see what other repos are starred by people who starred the same project. Great for exploring new domains. Generated by ChatGPT, actually. Showcase

I’ve built Cronus, a tool that makes it easier to write and understand cron expressions across different languages and time zones.

It shows human‑readable explanations of your cron jobs and adapts to multiple locales and time zones. You can preview schedules, tweak them visually, and copy/paste cron syntax for various environments.

I’d love feedback from folks who deal with cron jobs regularly—what’s missing, what would make it more powerful, and whether this solves any pain points you’ve had.

I recently went on a bike ride that felt much harder than it should have. Lo and behold, the primary culprit was the wind (and my legs). So, with Claude's help, I built a tool that visualizes wind conditions based on position and time using GPX file data and historical weather information. Maybe other people find it interesting/useful :)

Hi HN

I wanted a way to Q&A with educational channels I frequently watch e.g. personal development, coding, or building startups. But didn't want to go through the hassle of manually searching for the relevant content for the questions I had.

I wanted a tool to help me grow on YouTube by analyzing competitors to spot patterns in their content and see what makes them perform so well, and repurpose it for my channel.

And sometimes I just wanted to recall something I heard before in a podcast, but didn't want to go through the entire channel or video to get the answer.

So I built transcribr.io.

How it works: > Enter video URL or channel handle. > Video transcript(s) get extracted and processed for RAG. > Ask anything about the video or channel. > Get answers with exact video/timestamp references.

Now I can essentially chat with entire YouTube channels as individual knowledgebases in just a few clicks.

For example, I've been using it a lot to chat with the Y Combinator YouTube channel. Instead of watching hundreds of hours of startup school videos, office hours, and founder interviews, I just ask my specific questions and get back answers with exact timestamps from across their entire video library.

Try it here: https://transcribr.io $19.99/week or $49.99/mo - unlimited usage

Would love to hear about your use cases for such a tool.

This is my first ever submission to HN. Hope to receive some constructive criticism. Plan to submit more.

ZapHunt is an interactive audio scavenger hunt, where in each audio episode leads to another after finding an item in your home. No sign up needed.

This idea came from my journey as a parent — my daughter used to be overstimulated by noisy toys and screen time (not much anymore). I imagined as my kid grew older, I would want to find something to entertain without too much screen involvement. I noticed everyone liked scavenger hunts but they are like TODO lists, they get boring after a certain point of time. So, I combined it with a twist. Scavenger hunt with stories. Although, I intended for kids ages 7+, If you are a kid at heart you can try and give me feedback.

I’d love feedback from the community on a few things:

Did you want to play another after completing the whole story ?

Did the audio interactions feel smooth or confusing?

Should I pivot ZapHunt’s audio scavenger hunt concept to a new format?

Are there any industries or areas where these kinds of interactive audio experiences could be useful? I usually lean towards edutainment, but I'm wondering if there are other niche areas that might benefit from this idea, especially ones with fewer restrictions/laws around it ?

Any bugs, friction points, or parts that felt off?

Hi HN!

We’re building Omnicode at @leftlabs_ – a developer tool that lets you create complete Next.js, Vite, or Expo projects with just a single natural language prompt.

The idea is simple: skip boilerplate, setup, and config. Just describe what you want – e.g., “a blog with dark mode, markdown support, and a responsive navbar” – and Omnicode spins up the whole codebase in minutes.

Key Features:

Supports Next.js, Vite, and React Native (Expo)

Instant project scaffolding via prompt

Minimal, clean code output

Built for solo devs, hackathon MVPs, and fast prototyping

We’re inspired by tools like @lovable_dev and wanted to build something India-native, optimized for speed and simplicity.

We’d love your feedback, ideas, and brutal honesty.

Try it here We’re also open to collabs, integrations, or helping open-source builders.

Thanks! – Mandar (@leftlabs_)

Hi HN,

I built https://kocial.net, a platform that enables anyone to create their own branded community space, complete with features such as posts, reactions, and comments, similar to Hacker News, all without needing to write a single line of code.

You can use your own Custom domain as well for your community, like the community below:

- Indie Hackers - https://kocial.co

You can also create "invite-only" and "private" communities for your own private customer or fan base. I hope this helps.

I’d love feedback on the UX, performance, or any aspect of the niche.