Media is too big
VIEW IN TELEGRAM
π¦ hivepaas/shellwing
shellwing
Open encrypted server connections from your phone, laptop, and desktop, all synced through one secure vault. shellwing is a fast, native app for Mac, Windows, and Linux that turns a messy pile of saved passwords and scattered connection details into a single, organized tool. It handles the tricky parts of remote work for you, like keeping files in sync and managing keys automatically. The most interesting part is that your notes and secrets travel with you, encrypted end-to-end, so you can switch devices without re-typing a thing. It is a calm, powerful way to stay connected to your servers without the usual headache.
π° https://news.ycombinator.com/item?id=50184407
π @hackernewsgithubprojects
shellwing
Open encrypted server connections from your phone, laptop, and desktop, all synced through one secure vault. shellwing is a fast, native app for Mac, Windows, and Linux that turns a messy pile of saved passwords and scattered connection details into a single, organized tool. It handles the tricky parts of remote work for you, like keeping files in sync and managing keys automatically. The most interesting part is that your notes and secrets travel with you, encrypted end-to-end, so you can switch devices without re-typing a thing. It is a calm, powerful way to stay connected to your servers without the usual headache.
π° https://news.ycombinator.com/item?id=50184407
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ kevintngo/looseleaf
Looseleaf: The React Library That Looks Like Hand-Drawn Paper
Looseleaf is the React component library that makes your digital interface look like a hand-sketched paper notebook. It provides one hundred and twenty-three accessible components that mimic the texture of paper, ink, and tape, while still working like a standard tool. This solves the problem of sterile, flat user interfaces by adding a warm, tactile feel without sacrificing functionality. It is interesting because it proves you can have personality and polish in code simultaneously. The takeaway is that design doesn't always have to be sleek and corporate; sometimes, it just needs to feel human.
π @hackernewsgithubprojects
Looseleaf: The React Library That Looks Like Hand-Drawn Paper
Looseleaf is the React component library that makes your digital interface look like a hand-sketched paper notebook. It provides one hundred and twenty-three accessible components that mimic the texture of paper, ink, and tape, while still working like a standard tool. This solves the problem of sterile, flat user interfaces by adding a warm, tactile feel without sacrificing functionality. It is interesting because it proves you can have personality and polish in code simultaneously. The takeaway is that design doesn't always have to be sleek and corporate; sometimes, it just needs to feel human.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ andrewfader/blokkily
Blokkily: The AI-Powered Pattern DAW
Compose music by talking to your DAW. Blokkily is an experimental digital audio workstation that lets you build songs using large language models and agent control. Instead of just clicking buttons, you can ask the integrated assistant for musical phrases, have it generate them, and then review the results before applying them to your tracks. It also features a unique pattern-based workflow, where you can edit the same musical idea in a tracker, piano roll, or step grid, and see the changes instantly without stopping playback.
π° https://news.ycombinator.com/item?id=50064438
π @hackernewsgithubprojects
Blokkily: The AI-Powered Pattern DAW
Compose music by talking to your DAW. Blokkily is an experimental digital audio workstation that lets you build songs using large language models and agent control. Instead of just clicking buttons, you can ask the integrated assistant for musical phrases, have it generate them, and then review the results before applying them to your tracks. It also features a unique pattern-based workflow, where you can edit the same musical idea in a tracker, piano roll, or step grid, and see the changes instantly without stopping playback.
π° https://news.ycombinator.com/item?id=50064438
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ rerun-build/rerun
Rerun: AI Coworkers That Run Your Work
Rerun is the open source platform that turns AI agents into a coordinated team of virtual coworkers. Unlike standard chatbots, these agents work on a schedule, remember your preferences, and check in before taking sensitive actions. The standout feature is the Foreman, a supervisor agent that interviews you to understand your goals, then automatically hires and manages specialized coworkers to handle tasks like email sorting or lead generation. You get a visual board to monitor their progress, and they can use your existing tools like Gmail, Slack, and Notion. Because it is self-hostable, you retain full control over your data and workflows.
π @hackernewsgithubprojects
Rerun: AI Coworkers That Run Your Work
Rerun is the open source platform that turns AI agents into a coordinated team of virtual coworkers. Unlike standard chatbots, these agents work on a schedule, remember your preferences, and check in before taking sensitive actions. The standout feature is the Foreman, a supervisor agent that interviews you to understand your goals, then automatically hires and manages specialized coworkers to handle tasks like email sorting or lead generation. You get a visual board to monitor their progress, and they can use your existing tools like Gmail, Slack, and Notion. Because it is self-hostable, you retain full control over your data and workflows.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ cygaco/master-console
Master Console: Your AI Code Army
Turn a single AI assistant into a disciplined software company. Master Console installs a full organizational structure directly into your project, giving one chat interface the power of an entire engineering team. Instead of just coding, it spawns an architect that plans the work, builders that code in isolation, and strict reviewers that check every line of code before it moves forward. It even remembers what it learned in previous sessions so it does not make the same mistakes twice. This solves the chaos of solo development by adding the structure of a real team to your workflow.
π @hackernewsgithubprojects
Master Console: Your AI Code Army
Turn a single AI assistant into a disciplined software company. Master Console installs a full organizational structure directly into your project, giving one chat interface the power of an entire engineering team. Instead of just coding, it spawns an architect that plans the work, builders that code in isolation, and strict reviewers that check every line of code before it moves forward. It even remembers what it learned in previous sessions so it does not make the same mistakes twice. This solves the chaos of solo development by adding the structure of a real team to your workflow.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ vixhal-baraiya/qwengpu
qwenGPU: A GPU Inside a Language Model
qwenGPU is the experimental model that actually runs real GPU code inside its own brain. It takes the Qwen language model and physically compiles a tiny graphics processor directly into its neural network weights. The wild part is that when the model does math, it doesnβt just guess the answer. It writes out every single step of a graphics kernel, token by token, executing the actual code. Itβs less like a calculator and more like watching a computer think through a hardware pipeline. If you like watching machines do things that should be impossible, this is the project to look at.
π @hackernewsgithubprojects
qwenGPU: A GPU Inside a Language Model
qwenGPU is the experimental model that actually runs real GPU code inside its own brain. It takes the Qwen language model and physically compiles a tiny graphics processor directly into its neural network weights. The wild part is that when the model does math, it doesnβt just guess the answer. It writes out every single step of a graphics kernel, token by token, executing the actual code. Itβs less like a calculator and more like watching a computer think through a hardware pipeline. If you like watching machines do things that should be impossible, this is the project to look at.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ earendil-works/djs
DJS: The JavaScript Engine That Fits in a Pocket
DJS is the tiny JavaScript engine that treats your code like a frozen block of ice. Built entirely in Rust without any operating system dependencies, it packs the entire runtime into a single byte arena. This means you can snapshot your exact machine state by simply copying bytes, then restore it instantly on any machine. It was designed specifically to run scripts inside artificial intelligence agents, where speed and determinism matter more than everything else. By removing the usual overhead of standard libraries and complex memory management, DJS starts up in mere microseconds. It is not just a faster interpreter, but a completely different way to think about executing code.
π @hackernewsgithubprojects
DJS: The JavaScript Engine That Fits in a Pocket
DJS is the tiny JavaScript engine that treats your code like a frozen block of ice. Built entirely in Rust without any operating system dependencies, it packs the entire runtime into a single byte arena. This means you can snapshot your exact machine state by simply copying bytes, then restore it instantly on any machine. It was designed specifically to run scripts inside artificial intelligence agents, where speed and determinism matter more than everything else. By removing the usual overhead of standard libraries and complex memory management, DJS starts up in mere microseconds. It is not just a faster interpreter, but a completely different way to think about executing code.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ sjtu-deng-lab/waveforcing
Wave Forcing: Real-Time AI Video Generation
Wave Forcing generates a hundred seconds of high-definition video in under twelve seconds using a single cluster of eight graphics cards. This project solves the problem of making AI video generation fast enough to stream in real time. It achieves this by breaking the complex task of creating images into smaller chunks that different parts of the computer handle simultaneously. This parallel approach eliminates the waiting time that usually slows these systems down. The result is a video generator that runs at nearly the speed of light, making high-quality AI video accessible for live applications rather than just offline rendering.
π @hackernewsgithubprojects
Wave Forcing: Real-Time AI Video Generation
Wave Forcing generates a hundred seconds of high-definition video in under twelve seconds using a single cluster of eight graphics cards. This project solves the problem of making AI video generation fast enough to stream in real time. It achieves this by breaking the complex task of creating images into smaller chunks that different parts of the computer handle simultaneously. This parallel approach eliminates the waiting time that usually slows these systems down. The result is a video generator that runs at nearly the speed of light, making high-quality AI video accessible for live applications rather than just offline rendering.
π @hackernewsgithubprojects
Media is too big
VIEW IN TELEGRAM
π¦ astding/mmaemu
MMAEMU: Fixing GPU Math Mismatches
Make your AI model run identically on different graphics cards, even when the hardware is fundamentally different. That is the promise of Mmaemu, a tool that recreates the exact math of high-end NVIDIA chips on your standard computer. The problem it solves is subtle: when you run modern language models, different graphics cards perform tiny floating-point calculations slightly differently. Usually, those tiny differences vanish, but sometimes they cause your results to drift wildly off track. Mmaemu steps in by emulating the precise bit-level arithmetic of reference chips.
π @hackernewsgithubprojects
MMAEMU: Fixing GPU Math Mismatches
Make your AI model run identically on different graphics cards, even when the hardware is fundamentally different. That is the promise of Mmaemu, a tool that recreates the exact math of high-end NVIDIA chips on your standard computer. The problem it solves is subtle: when you run modern language models, different graphics cards perform tiny floating-point calculations slightly differently. Usually, those tiny differences vanish, but sometimes they cause your results to drift wildly off track. Mmaemu steps in by emulating the precise bit-level arithmetic of reference chips.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ globalmpc/nomore
Stop Repeating Yourself to Claude Code
nomore reads your local conversation history to spot the exact instructions you have to repeat over and over again, then turns those repeated corrections into permanent rules that save your time. It works without sending your data to a cloud server, so you keep full control. The project also includes a pre-built collection of twenty common fixes, mined directly from thousands of public bug reports, that you can install in seconds. This means you no longer have to manually type out the same preferences for every new session, which makes working with the AI much faster and less frustrating.
π° https://news.ycombinator.com/item?id=50274288
π @hackernewsgithubprojects
Stop Repeating Yourself to Claude Code
nomore reads your local conversation history to spot the exact instructions you have to repeat over and over again, then turns those repeated corrections into permanent rules that save your time. It works without sending your data to a cloud server, so you keep full control. The project also includes a pre-built collection of twenty common fixes, mined directly from thousands of public bug reports, that you can install in seconds. This means you no longer have to manually type out the same preferences for every new session, which makes working with the AI much faster and less frustrating.
π° https://news.ycombinator.com/item?id=50274288
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ vishakhasadhwani/devops-ai-playbook
DevOps AI Playbook
Turn your terminal into a smart command center where an AI assistant doesn't just write code but actively manages your live cloud infrastructure. This project builds a real microservices app on Amazon Web Services while wiring up an intelligent agent named Kira to handle the messy work of monitoring and troubleshooting. The coolest part is how it connects the dots: you deploy the system, and the AI reads the logs, checks the cluster health, and helps you fix issues in real time. Itβs less about memorizing commands and more about learning how to let intelligent tools handle the heavy lifting.
π @hackernewsgithubprojects
DevOps AI Playbook
Turn your terminal into a smart command center where an AI assistant doesn't just write code but actively manages your live cloud infrastructure. This project builds a real microservices app on Amazon Web Services while wiring up an intelligent agent named Kira to handle the messy work of monitoring and troubleshooting. The coolest part is how it connects the dots: you deploy the system, and the AI reads the logs, checks the cluster health, and helps you fix issues in real time. Itβs less about memorizing commands and more about learning how to let intelligent tools handle the heavy lifting.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ vocahq/vocamac
VocaMac: Private Voice Typing for Mac
VocaMac is the private dictation tool that finally lets you speak into your Mac without sending a single word to the cloud. You simply hold down a key, talk, and the text appears right at your cursor in whatever app you are using, like your email or code editor. The coolest part is that it runs entirely on your computer using local speech engines, meaning your conversations stay strictly on your device with no account or subscription needed. It even cleans up your filler words automatically so you get polished text without the manual editing work.
π @hackernewsgithubprojects
VocaMac: Private Voice Typing for Mac
VocaMac is the private dictation tool that finally lets you speak into your Mac without sending a single word to the cloud. You simply hold down a key, talk, and the text appears right at your cursor in whatever app you are using, like your email or code editor. The coolest part is that it runs entirely on your computer using local speech engines, meaning your conversations stay strictly on your device with no account or subscription needed. It even cleans up your filler words automatically so you get polished text without the manual editing work.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ cglide/comfyui-cglide
CGlide for ComfyUI
CGlide stops your video clips from looking like they were baked in the oven. When you link short clips together to make a longer film, the usual method degrades the image quality with every connection, making it duller and crunchier. This tool fixes that by keeping the raw data intact instead of converting it to pixels and back. It also lets you lock a real song to the footage, so the animation actually matches the beat instead of guessing the music. It is a complete workflow for making short films where the colors stay perfect from start to finish.
π @hackernewsgithubprojects
CGlide for ComfyUI
CGlide stops your video clips from looking like they were baked in the oven. When you link short clips together to make a longer film, the usual method degrades the image quality with every connection, making it duller and crunchier. This tool fixes that by keeping the raw data intact instead of converting it to pixels and back. It also lets you lock a real song to the footage, so the animation actually matches the beat instead of guessing the music. It is a complete workflow for making short films where the colors stay perfect from start to finish.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ leonxlnx/golden-temple
Golden Temple: 24 Million Triangles From Code
Twenty-four million triangles, zero texture files. That is the reality of Golden Temple, a project that constructs an entire golden-hour Japanese temple complex using thirty-one thousand lines of JavaScript. No pre-made 3D models and no downloaded images exist here; every roof tile, lotus flower, and koi fish is generated mathematically in real time. The single most striking feature is the final output: a seventy-second cinematic film rendered offline frame by frame. It proves you can build a photorealistic world entirely from logic. If you want to see how far procedural generation can go, this is the benchmark.
π @hackernewsgithubprojects
Golden Temple: 24 Million Triangles From Code
Twenty-four million triangles, zero texture files. That is the reality of Golden Temple, a project that constructs an entire golden-hour Japanese temple complex using thirty-one thousand lines of JavaScript. No pre-made 3D models and no downloaded images exist here; every roof tile, lotus flower, and koi fish is generated mathematically in real time. The single most striking feature is the final output: a seventy-second cinematic film rendered offline frame by frame. It proves you can build a photorealistic world entirely from logic. If you want to see how far procedural generation can go, this is the benchmark.
π @hackernewsgithubprojects
Media is too big
VIEW IN TELEGRAM
π¦ zeke/faster-chrome-devtools-skill
faster chrome devtools skill: the agent skill that drives Chrome without the heavy stuff
faster chrome devtools skill is the agent skill that drives Chrome without the heavy stuff. It lets your coding agent control a real browser directly, skipping the usual heavy setup. You don't need to install any extra packages or configure servers. The tool connects straight to Chrome, so you can open pages, click buttons, and grab data instantly. It is lean and fast because it uses the browser's native protocol. No more waiting for bloated automation tools to boot up. Just plain, direct control over your web session. It is the quickest way to let your code browse the web. You will love how simple it feels.
π @hackernewsgithubprojects
faster chrome devtools skill: the agent skill that drives Chrome without the heavy stuff
faster chrome devtools skill is the agent skill that drives Chrome without the heavy stuff. It lets your coding agent control a real browser directly, skipping the usual heavy setup. You don't need to install any extra packages or configure servers. The tool connects straight to Chrome, so you can open pages, click buttons, and grab data instantly. It is lean and fast because it uses the browser's native protocol. No more waiting for bloated automation tools to boot up. Just plain, direct control over your web session. It is the quickest way to let your code browse the web. You will love how simple it feels.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ phone-harness/mcp
Your AI Agent Just Grabbed a Real iPhone
Your AI agent can now physically grab a real iPhone and start tapping, swiping, and typing on it like you. This project, called Phone Harness, connects your favorite AI tools to actual cloud phones, so they can see the screen, launch apps, and navigate the interface just like a human finger would. It solves the problem of agents being stuck in text-only chat boxes by giving them hands on real devices. You do not need to buy hardware or set up complex cables; just paste a link into your AI client, sign in, and the agent starts driving the phone.
π @hackernewsgithubprojects
Your AI Agent Just Grabbed a Real iPhone
Your AI agent can now physically grab a real iPhone and start tapping, swiping, and typing on it like you. This project, called Phone Harness, connects your favorite AI tools to actual cloud phones, so they can see the screen, launch apps, and navigate the interface just like a human finger would. It solves the problem of agents being stuck in text-only chat boxes by giving them hands on real devices. You do not need to buy hardware or set up complex cables; just paste a link into your AI client, sign in, and the agent starts driving the phone.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ twigpine/openrecall
OpenRecall: AI Memory That Stays Private
OpenRecall is the memory layer that ensures your AI agents never forget what they know, even if you switch tools. It solves the problem of losing context by creating one encrypted memory bank that works across different coding assistants and projects. The surprising part is that the server only ever sees scrambled data; the actual notes are locked on your machine, so no one else can read them. It uses a standard connection protocol to let any compatible agent save and recall durable facts, meaning your preferences and project decisions stick around between sessions. You can run it locally or host it yourself to sync across your computers.
π @hackernewsgithubprojects
OpenRecall: AI Memory That Stays Private
OpenRecall is the memory layer that ensures your AI agents never forget what they know, even if you switch tools. It solves the problem of losing context by creating one encrypted memory bank that works across different coding assistants and projects. The surprising part is that the server only ever sees scrambled data; the actual notes are locked on your machine, so no one else can read them. It uses a standard connection protocol to let any compatible agent save and recall durable facts, meaning your preferences and project decisions stick around between sessions. You can run it locally or host it yourself to sync across your computers.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ huwhitememes/comfyui-krea2-conditioning
Fix Your AI Faces: Krea 2 Conditioning Control
ComfyUI Krea 2 Conditioning Control is the custom node that finally stops your AI portraits from developing weird skin spots and shifting facial features. The previous method tried to boost image detail by amplifying the text signal, but that brute-force approach caused the model to drift, creating artifacts and making faces look younger or tilted. This tool fixes it by rebalancing the internal layers while keeping the total signal strength exactly the same. It uses a smart math trick to preserve the overall volume, so you get sharper details without the quality collapse.
π @hackernewsgithubprojects
Fix Your AI Faces: Krea 2 Conditioning Control
ComfyUI Krea 2 Conditioning Control is the custom node that finally stops your AI portraits from developing weird skin spots and shifting facial features. The previous method tried to boost image detail by amplifying the text signal, but that brute-force approach caused the model to drift, creating artifacts and making faces look younger or tilted. This tool fixes it by rebalancing the internal layers while keeping the total signal strength exactly the same. It uses a smart math trick to preserve the overall volume, so you get sharper details without the quality collapse.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ storytold/materialcraft
MaterialCraft: The Rust Material Designer
A brick wall, a lava rock, and polished marble are all built from math, not photos, inside MaterialCraft. This open-source tool lets you design surfaces by connecting nodes in a graph, so you can create textures that tile perfectly without ever needing a photograph. The most surprising part is that the entire engine is written in pure Rust, meaning it runs natively on your computer or directly in your browser without any heavy web wrappers. It is currently in active development, offering a familiar workspace for artists who want total control over their materials.
π @hackernewsgithubprojects
MaterialCraft: The Rust Material Designer
A brick wall, a lava rock, and polished marble are all built from math, not photos, inside MaterialCraft. This open-source tool lets you design surfaces by connecting nodes in a graph, so you can create textures that tile perfectly without ever needing a photograph. The most surprising part is that the entire engine is written in pure Rust, meaning it runs natively on your computer or directly in your browser without any heavy web wrappers. It is currently in active development, offering a familiar workspace for artists who want total control over their materials.
π @hackernewsgithubprojects
This media is not supported in your browser
VIEW IN TELEGRAM
π¦ dissimilis/flashtester
FlashTester: Exposing Fake USB Drives
FlashTester is the diagnostic tool that exposes fake USB drives in minutes rather than hours. It solves the problem of counterfeit flash drives that report a massive storage capacity but actually hold only a fraction of that space. Unlike older tools that write the entire drive from start to finish, this software uses a smart sampling technique to find the real data boundary almost instantly. If a drive is fake, it shows you exactly where the storage ends and offers to reformat it to that true size so your files stop disappearing.
π @hackernewsgithubprojects
FlashTester: Exposing Fake USB Drives
FlashTester is the diagnostic tool that exposes fake USB drives in minutes rather than hours. It solves the problem of counterfeit flash drives that report a massive storage capacity but actually hold only a fraction of that space. Unlike older tools that write the entire drive from start to finish, this software uses a smart sampling technique to find the real data boundary almost instantly. If a drive is fake, it shows you exactly where the storage ends and offers to reformat it to that true size so your files stop disappearing.
π @hackernewsgithubprojects