2025年9月8日 の Show HN

Hi HN,

I have created a webpage that displays all C++ features since C++20 in a simple, searchable table.

It is intended to serve as a quick reference for C++ developers, whether as support for cross-platform development or simply to track the current support status out of curiosity.

I created it as a simpler, more structured, and more up-to-date alternative to the cppreference compiler support site. Please note that the page intentionally does not list LWG and CWG papers. This might change as I am continually updating the site and trying out new ideas.

Questions, feedback and suggestions are appreciated, either here or in the form of GitHub issues.

We're happy to present Veena Chromatic Tuner, an app we've developed for musicians, instrument makers, and ethnomusicologists who need more than just a standard chromatic tuner. Our goal was to create a tool that not only supports pitch detection but also provides deep support for diverse musical intonation systems and offers intuitive visual feedback.

The Problem We're Solving: Many tuners are good for Equal Temperament, but has limited support when it comes to the Just Intonation, microtonal music, or the specific requirements of instruments like the Veena where fret positions are determined by precise ratios.

Oscilloscope-like Visual Feedback: Instead of just a needle, you get a dynamic, oscilloscope-like waveform display.

In Tune: The waveform appears stabilized, giving you an immediate, confirmation of perfect pitch. Sharp: The waveform rotates right. Flat: The waveform rotates left.

This dynamic visual feedback, akin to a digital oscilloscope's trigger synchronization, offers immediate, precise adjustment cues that go far beyond what a static needle can provide, allowing for incredibly fine-tuned adjustments.

Unmatched Intonation Flexibility: We understand that music isn't just 12-TET.

Just Intonation Support: Perfect for Indian classical music, early music, and any tradition that relies on pure harmonic relationships between notes. This is crucial for achieving the rich, resonant chords and melodic purity that Equal Temperament can't always deliver.

Custom Temperaments: Go beyond presets! Create, save, and manage your own unique temperaments with personalized ratio settings. This empowers composers, researchers, and performers to explore microtonal scales and historical tunings with ease.

Dedicated Veena Instrument Mode: It allows users to play and tune notes across 24 fret positions, specifically highlighting how note positions on the fretboard vary relative to each other when pure intonation is applied. This feature is invaluable for instrument makers and those studying the physics of string instruments.

Other Key Features:

Multicultural Note Naming: Display notes in Western, Indian classical (Carnatic/Hindustani), and Solfege, with support for multiple Indian language scripts (Tamil, Devanagari, Telugu, Kannada, Malayalam).

Adjustable Reference Pitch: Customize A4 frequency from 440Hz to 432Hz or anything in between.

Transposition Support: Easily transpose notes for different instruments.

This app is would be useful for string players (veena, violin, guitar, sitar, etc.), wind instrument musicians, vocalists, music teachers, students, ethnomusicologists, and especially instrument makers and tuners who need to work with precise intonation and fret setting. Anyone exploring microtonal music will also find it incredibly useful.

We're actively developing the app, continuously adding features and improvements. While pitch detection is resource-intensive, we strive for broad device compatibility.

Check it out on Google Play: [https://play.google.com/store/apps/details?id=in.magima.digi...]

We welcome your feedback, questions, and thoughts!

Just a toy, showing how easy it is to leverage built-in Emacs features (most notably Artist mode, which provides a set of functions for creating ASCII-art vector graphics) and things like trigonometric functions and timers to create something nice.

A short blog post mentioning some background: https://mbork.pl/2025-09-08_Emacs_Artist_clock#.

We wanted to run multiple AI coding assistants in parallel and swap models easily (Claude Code, local, Codex, etc.) without messing with the current branch.

DevSwarm is a mac/windows desktop app that runs assistants on separate git branches so you can stay in the loop and compare/merge safely. Not an IDE, an Augmented Development Environment (ADE); open any branch in your IDE with a click. We’ve been dogfooding it for months, including DevSwarm. Try it in minutes: download, open a repo, start two assistants. Free beta.

A online interactive map Explore Hollow Knight: Silksong’s world with Silksong Map – bosses, benches, collectibles, and fast travel.

Hey! So I worked on a new website for my TUI called `dash` on https://gh-dash.dev

The docs content was contributed by a community member @michaeltlombardi https://github.com/michaeltlombardi which I'm super grateful for.

Also, for any community member I created a discord server (https://discord.com/invite/SXNXp9NctV) where we can share configs and nerd out about terminal.

The site itself is built with Starlight by Astro. I've set up a VPS with Dokploy (https://dokploy.com/) which made things so easy! Each PR publishes a new version of the site.

I love ASCII art so I sprinkled just a tiny bit (still need to practice :D).

Let me know what you think!

I'm an infrastructure architect who started using AI assistants to write code 3 months ago. After building several systems with Claude, I noticed a pattern: the code always had security issues I could spot from my ops background, but I couldn't fix them myself since I can't actually write code.

Why I built this: I needed a way to verify AI-generated code was production-safe. Existing tools either required cloud uploads (privacy concern) or produced output too large for AI context windows. TheAuditor solves both problems - it runs completely offline and chunks findings into 65KB segments that fit in Claude/GPT-4 context limits.

What I discovered: Testing on real projects, TheAuditor consistently finds 50-200+ vulnerabilities in AI-generated code. The patterns are remarkably consistent: - SQL queries using f-strings instead of parameterization - Hardcoded secrets (JWT_SECRET = "secret" appears in nearly every project) - Missing authentication on critical endpoints - Rate limiting using in-memory storage that resets on restart

Technical approach: TheAuditor runs 14 analysis phases in parallel, including taint analysis (tracking data from user input to dangerous sinks), pattern matching against 100+ security rules, and orchestrating industry tools (ESLint, Ruff, MyPy, Bandit). Everything outputs to structured JSON optimized for LLM consumption.

Interesting obstacle: When scanning files with vulnerabilities, antivirus software often quarantines our reports because they contain "malicious" SQL injection patterns - even though we're just documenting them. Had to implement pattern defanging to reduce false positives.

Current usage: Run aud full in any Python/JS/TS project. It generates a complete security audit in .pf/readthis/. The AI can then read these reports and fix its own vulnerabilities. I've seen projects go from 185 critical issues to zero in 3-4 iterations.

The tool is particularly useful if you're using AI assistants for production code but worry about security. It provides the "ground truth" that AI needs to self-correct.

Would appreciate feedback on: - Additional vulnerability patterns common in AI-generated code - Better ways to handle the antivirus false-positive issue - Integration ideas for different AI coding workflows

Thanks for taking a look! /TheAuditorTool

I spent the weekend working with Claude Code to build out a CLI script that can take a google sheet that has input tabs, output tabs, and processing tabs (which is the way I tend to build them), and write a python or typescript/javascript file that will do the same logic as the google sheet does. The general use case for this is allowing non-software-developers (e.g., finance professionals, actuaries, scientists) to build out data processing logic, and then be able to run it more quickly/effectively than using the google sheets API.

I'm not sure how niche this is, and I am still testing it out with ever-more-complicated sheets, but so far it's been surprisingly effective for me. Happy to get any comments/thoughts/suggestions/etc.

I built an iOS voice assistant that connects your action button to Gemini Live with 18 native iOS tools like location, calendar, and so on. It also connects to any shortcuts you have on your phone. Totally free, no account, no setup.

I put together a cybersecurity quiz covering most major topics. It works as both a learning tool and an easy way to test your knowledge.

Some areas have lighter coverage due to cybersecurity's broad scope.

Let me know if you find questions that need better answers.

Enjoy!

Tony here.

Over the past year (and post covid), more and more companies have been rolling back their remote work policies:

- 75% of US employees were required to be in-office regularly in late 2024 (up from 63% in 2023). - 48% of UK employers now expect full on-site attendance, compared to just 27% in 2023.

But the thing is employees still overwhelmingly want flexible or fully remote work.

I can't think of anything more soul destroying than losing hours every day sitting in traffic just to get to and from the office.

Here's the good news: plenty of companies are still committed to remote work. The hard part is figuring out who they are.

So that's why I built https://www.remoteweek.io. It's a job board that only lists remote roles.

If you’re interested in checking out some remote jobs/companies (or want to share feedback), you can check it out here: https://www.remoteweek.io.

And if you know of any remote-first companies sticking to their guns, let me know. In the meantime I’ll keep filling the database with more jobs.

After months of development, I'm excited to release uprintf v1.3 - the ultimate platform-independent printf replacement that works everywhere from bare-metal microcontrollers to modern OSes.

What's new in v1.3:

-Template System - Now with {{variable}} substitution like modern template engines -Multi-Output Streaming - Broadcast to multiple destinations simultaneously (UART, files, network) -String Builder - Dynamic string building with efficient memory management -Text Transformation - Built-in uppercase, lowercase, capitalize, reverse, and ROT13 transforms -Advanced Formatting - Text alignment, wrapping, and pattern-based formatting -Event System - Hook into output streams for monitoring and processing -Stream Processors - Real-time character transformation pipelines -State Machine - Lightweight state management for complex workflows

And it's still: -Zero dependencies -Header-only -Fully portable -MIT licensed

Perfect for embedded systems, CLI tools, and anywhere you need powerful formatting without bloat.

GitHub: https://github.com/Ferki-git-creator/Uprintf

Hi HN,

I built GoSocket, a small library that helps you set up WebSocket servers in Go with almost no boilerplate.

It lets you start a server in a few lines of code, handle connections, broadcast to rooms or clients, and add middleware. JSON is supported now, Protobuf and MessagePack are planned.

Example:

ws := gosocket.NewServer() ws.WithPort(8080). WithPath("/ws"). OnMessage(func(c gosocket.Client, m gosocket.Message, ctx *gosocket.HandlerContext) error { c.Send(m.RawData) // echo return nil }) log.Fatal(ws.Start())

Repo: https://github.com/FilipeJohansson/gosocket

It’s still pre-1.0, so I’d love feedback, testing, and contributions.

Hey everyone,

I just launched Astrae v2, a library of modern Next.js landing page templates and animated components.

The goal is to make it easier for developers and designers to build beautiful, production-ready sites fast, without spending hours on layout, animations, or responsiveness.

What’s new in v2:

• More templates (landing pages, SaaS, portfolio, etc.) • Polished UI with Framer Motion animations • Design Figma files • Copy-paste ready components (cards, hero sections, pricing tables, etc.) • Improved code quality for easier customization

This started as a personal project for my client work, but it’s grown into something I wanted to share with the community. I’d love your feedback, what works, what doesn’t, and what you’d like to see in future updates.

To celebrate launch day, I’m running a 20% discount (today only) Use code CF59AP at checkout.

Thanks for checking it out.

Yo! Messed around this weekend with creating a good interface for nano banana (which I think is the best image editing model out there at the moment, and Google's product interfaces leave a lot to be desired).

Use code MOCKMIKELOVESHN for a free month of the Starter membership if you wanna mess around with it.

Cheers!

After trying to find a ui Library which works which works well with Expo and react native web. I decided to create my own.

Cross platform: iOS, Andriod, and the web with Windows and macOS support on the way.

Our components are accessible on Native and the web. To achieve web accesibility we use react-aria

This is what differentiates us even beyond shadcn/ui! Our components include extra variants so you can adjust error, loading, borderRadius, and style without writing any extra tailwind.

Storybook: We have included a storybook to let you play with all the variants.

Tested: We write unit tests for all the components to ensure they are working as expected.

Customizable: You own the code and the implementation. You can customize anything you want.

Get a live demo with storybook here https://main--68af7af75f72e5521a7e0f93.chromatic.com/?path=/...

Proxied Web from URL (https://web.818233.xyz) —— A free web proxy service directly from URL

Usage: Append any web URL to https://web.818233.xyz/, and the new URL will return the web content of the original URL.

I built this service mainly to extract content of any website for the use of LLM. It can also be handy if you do not have access to certain websites.

What if Pokemon were actually smart?

Howdy, I’m Ramon - and that’s exactly what my studio is working on! We’re making Allyon: Battlecry, a monster trainer game with lifelike creatures that know how to think and act on their own, like they were ripped straight out of an anime!

What makes it different: • Prepare: train your monster through voice and demonstrations to come up with killer moves to use in battle • Improvise: physics-driven arenas that change as you fight - douse flames into steam, blast craters, reshape cover. • Bond: Allyon understand natural language, have personalities and memories, and your relationship grows over time. Between fights you can just hang out, talk, and play.

If that sounds like your thing:

Steam wishlist (huge help): https://store.steampowered.com/app/3987700/Allyon_Battlecry/ Join the Discord for playtests & feedback: https://tinyurl.com/allyonbattlecry

Thanks, and I’d love to hear your thoughts!

I implemented a plugin for Flutter using platform views to integrate native SwiftUI Liquid Glass components. Here is the backstory and instructions for trying it out.

We built Blink AI to keep your social accounts active without the daily grind. - Spin up an AI agent that learns from your docs or sites and writes in your voice - Generate & schedule posts for X and LinkedIn (fully autonomous or approval-first) - Our Dashboard surfaces real-time trends and news - See what works with built-in analytics - (Coming soon) Browser extension that lets you react to any content across the web—in your own voice

Looking for feedback on the core workflow and gaps you’d want filled

Pluely is an open-source alternative to all stealth AI products/companies, and Pluely is your go-to AI companion that lives silently on your desktop as a translucent overlay—always on display, one click away. At just ~10 MB, it’s ultra-lightweight (27× smaller than Cluely), uses 50% less compute power, and launches in < 100 ms. Pluely stays on top of any window without intrusion, overlaps seamlessly with your workflow, and requires zero configuration—just your API keys and you’re ready to summon AI insights, summaries, translations, and more in an instant.

Core Features: System Audio Capture: Transcribe any audio—Zoom calls, YouTube videos, podcasts, music, game commentary—in real time. Cross-platform support via BlackHole (macOS), Stereo Mix/VB-Cable (Windows), and PulseAudio monitors (Linux).

True Invisibility: No dock or taskbar icon. Always-on-top overlay with passthrough clicks. Global shortcuts for quick toggle and voice input. Undetectable in screen shares, recordings, and live streams.

Universal Provider System: Paste any cURL command; Pluely auto-parses it for full support of OpenAI-compatible or custom LLM/STT APIs, including Ollama, Gemini, Claude, Grok, Whisper, Deepgram, and more.

Local-First & Lightweight: Built with Tauri for native speed—just ~10 MB, launches in < 100 ms, uses 80% fewer resources than Electron alternatives. Runs entirely with your API keys; zero telemetry.

Why Pluely Exists: Existing “invisible” AI assistants are either heavy, closed-source, or require cloud backdoors. I built Pluely to be truly stealthy, privacy-first, and fully extensible—so you control your data and your AI.

What’s New in v0.1.3: Complete system audio pipeline with automatic speech detection

Enhanced stealth mode: vanishing icon, passthrough clicks, refined global shortcuts

Advanced cURL parsing for ANY provider, plus improved error handling

Premium API integration for unlimited usage and advanced STT (optional license)

Community Traction: [600+ stars on GitHub] and climbing

Get Started: Downloads: https://pluely.com/downloads Website: https://pluely.com latest v0.1.3 Release: https://github.com/iamsrikanthnani/pluely/releases/tag/app-v... Repo: https://github.com/iamsrikanthnani/pluely

You can add your own LLM provider and STT provider and use it for FREE with no limits, no signup. If you have a feature request or bug, just report it here: https://github.com/iamsrikanthnani/pluely/issues

Hell yeah! I had this idea since MCP came out, but I FINALLY got it to work: OpenAPI to MCP is a fact:

Convert ANY server described with openapi into an MCP endpoint!

mcp.openapisearch.com/anyhostname.tld/mcp

There are a few requirements to it, but: If the server serves REST described with OpenAPI and if it either has NO auth OR implements OAUTH the MCP way, it will automatically work!

Check the repo!

Hi HN,

We’ve been working on Hyperif, an AI assistant that connects directly to your tools (Google Calendar, Gmail, Slack, Ads, spreadsheets, etc.) so you can access data and take action — just by chatting.

The idea came from our customers. Over the years, we’ve had teams come to us who struggled with building workflows and never really understood APIs. They always needed help setting up their specific integrations. For most non-technical users, it’s the same issue: you end up asking someone else to “build this out” whenever you want to connect tools or automate something.

With AI making interaction so much simpler, we realized it doesn’t need to be this way. Instead of fiddling with APIs or dragging boxes in a workflow builder, what if you could just chat with your data?

With Hyperif you can:

- “What’s on my calendar tomorrow?” - “Summarize the google docs file” - “Pull data from this spreadsheet and send a quick summary to Slack.” - “Get ad insights and create a doc from the analysis.”

What’s different about Hyperif is that: - It connects via APIs (not UI hacks), so results are reliable. - You don’t need to set up automations or workflows — you just ask. - You can turn repeat prompts into “Agents” that run on-demand - It’s designed with security first (OAuth, no hidden data storage, enterprise options).

We’re already working with a few companies who connected not just their SaaS tools but also their bespoke internal software, which has been exciting to see.

You can try it here: https://hyperif.com

I’d love feedback on: - Which use cases resonate most (calendar, docs, email, ads, etc.) - Whether the “chat-first” approach feels simpler than building workflows - Any technical concerns or suggestions on integrations/security

This is an early release, so things may break — if they do, we’d really appreciate you letting us know.

Thanks!