📦 ajayvarmaramineni/byoc

BYOC: Your Files, Your Rules

BYOC is the storage library that lets your app save files directly to your own cloud instead of a company server. It unifies Google Drive, Nextcloud, and S3 into one simple interface, so you never pay hidden egress fees again. The coolest part is direct browser upload: users send files straight to their personal accounts without your server ever touching the data. Your app just sees a link. This keeps your infrastructure light and gives users total control over where their documents live. If you handle user files, this is the bridge you have been missing. Stop renting storage and start using what people already own.

🆔 @hackernewsgithubprojects

📦 jonjonrankin/pi-caveman

pi-caveman: Save 75% of Tokens

Cut your AI coding costs by up to seventy-five percent without losing a single bit of technical accuracy. Meet pi-caveman, a playful extension that forces your coding agent to speak like a caveman. Instead of wasting tokens on polite filler and long explanations, it demands short, punchy fragments. It’s like hiring a brilliant engineer who only has five seconds to talk, but somehow still solves your bugs perfectly. The result is a massive drop in output costs for the same quality work, making it a clever trick for anyone who cares about efficiency.

📰 https://news.ycombinator.com/item?id=47721514

🆔 @hackernewsgithubprojects

📦 halo-box/strix-llama.cpp

strix-llama.cpp

A single chip can hold nearly one hundred and twenty eight gigabytes of shared memory, letting you run massive artificial intelligence models locally without a server room. The Strix Llama C plus project is a specialized software fork built specifically to squeeze the maximum speed out of that hardware. It rewrites the low level math routines so they fit the unique quirks of the integrated graphics processor. This means you get faster text generation and better performance right on your desk. If you have this specific hardware, this is the only tool you need to unlock its true potential.

🆔 @hackernewsgithubprojects

📦 peonist-ai/halogen-flash-server

Halogen Flash Server: 4x Faster AI on One Chip

Halogen Flash Server makes running a massive, reasoning AI model on a single AMD chip feel instant. Most engines slow down badly as your conversation grows, but this one stays fast even with massive context. It achieves this by building custom math routines specifically for your hardware and model, skipping all the generic overhead that usually slows things down. The result is a server that processes long prompts in seconds where others take minutes. It is not just a quick demo; it is a fully working tool you can point your coding agent at right now.

🆔 @hackernewsgithubprojects

📦 vheissu/hit-and-run-web

Hit & Run Web: The Simpsons in Your Browser

This entire PS2 classic, The Simpsons: Hit & Run, now runs natively in your web browser without a single line of console emulation. The project is called Hit & Run Web, and it is a full reconstruction of the game using Three.js. You drive Bart, Homer, and the rest of the gang through seven original levels, racing and fighting the police with converted assets from a real disc image. The most impressive part is that it includes the actual story missions, bonus challenges, and even the interior shops you can walk into.

🆔 @hackernewsgithubprojects

📦 abrahampaulj/facefusion-mobile

FaceFusion Mobile: The Offline Android Face Swap App

A ten second video clip gets its faces swapped in thirteen seconds, entirely on your phone, without touching the cloud. FaceFusion Mobile is an Android application that runs the neural networks for face swapping directly on your device's specialized hardware, meaning no data ever leaves your pocket. If your phone lacks the specific chip for this, it gracefully steps down to standard processing to keep the job done. The most surprising part is that the output is identical to the powerful server version, yet it runs locally. This solves the privacy problem by ensuring your photos never get uploaded.

🆔 @hackernewsgithubprojects

📦 wsl043/dsh-codex-subscription

DSH Codex Subscription: Use Your ChatGPT Plan Inside DeepSeek Harness

DSH Codex Subscription lets you use your existing ChatGPT subscription directly inside DeepSeek Harness without needing a separate API key or command line tools. It acts as a bridge that connects your account to the platform, giving you access to various models, web search, and even image generation. The most useful part is that it keeps all your usage data and settings local, so you don't have to manage external credentials. It also includes a built-in viewer for generated images, allowing you to edit or download them right there.

🆔 @hackernewsgithubprojects

📦 pricootz/edgepilot

EdgePilot: The Screen Edge Monitor

A tiny pill sitting on the edge of your screen can pull out live data about your entire computer without needing a single cloud server. EdgePilot is a compact system monitor that lives on Windows and Linux, unfolding from the screen border to show you exactly how much CPU, memory, and disk space you are actually using right now. It feels like a hidden layer of your desktop, springing open with smooth motion only when you need to check what your machine is doing in the background. The whole project runs locally on your computer, meaning your data never leaves your device.

🆔 @hackernewsgithubprojects

📦 physical-superintelligence-lab/simple

SIMPLE: The Playground for Humanoid Robots

SIMPLE gives you a virtual playground where you can train humanoid robots to walk, bend, and grab things before they ever touch the real world. It solves the massive headache of testing complex movements on actual hardware by letting you run thousands of trials in a simulator, where mistakes are just code updates and not expensive repairs. The coolest part is that it acts like a universal adapter, letting you plug in almost any major robot brain and watch it perform in realistic room settings. So if you are building the next big robot, this is your safe space to get it right before you roll it out.

🆔 @hackernewsgithubprojects

📦 pinto0309/screen-eye-tracking

Screen Eye Tracking: Control Your PC With Just Your Eyes

Screen Eye Tracking is the desktop app that turns your webcam into a functional mouse pointer without any special hardware. It watches your face, finds your eyes, and calculates exactly where you are looking on your screen, dropping a red dot at that spot. The real magic happens when you enable the lip motion feature. Instead of fumbling for a mouse to click, you simply open and close your mouth to send a click command to wherever your gaze is hovering. It is a surprisingly accurate way to interact with your computer using nothing but your face and your webcam, proving that eye control does not always require expensive headsets.

🆔 @hackernewsgithubprojects

📦 achrefelouafi/linearabiltycastingextendedthreejs

LinearAbiltyCastingExtendedThreeJS: Freeze Time and Sculpt Your Magic

LinearAbiltyCastingExtendedThreeJS is a visual effects sandbox where you can freeze an explosion mid-air and keep tweaking it. Built with Three.js and hand-written shaders, it lets you pause a chaotic ability and adjust over two thousand live sliders without restarting. You can reshape crystals, change colors, and alter physics on a still frame, turning a static screenshot into an interactive playground. It is built for artists who want to see exactly how their magic looks without waiting for a full playback. If you like designing effects, this is your playground. Grab your coffee and start sculpting reality.

🆔 @hackernewsgithubprojects

📦 hust-open-atom-club/oh-dsh

One Runtime, Three Screens: The DSH Experience

Get your AI agent running on your desktop, in your browser, and straight from the terminal without writing a single line of glue code. That is exactly what oh-dsh does. It is a unified runtime that lets you build an agent once and see it live in three distinct places: a native desktop app, a web interface, and a text-based terminal. The problem it solves is the annoying reality that most developer tools are locked into one window type, forcing you to rebuild the same logic every time you switch screens.

🆔 @hackernewsgithubprojects

📦 timgordontg/engrim

engrim: The Save Button for Your AI Code

Your AI coding assistant has a terrible habit of forgetting everything the moment you clear the chat, which is basically like paying for a brilliant intern who amnesically loses their notes every afternoon. Engrim fixes this by acting as a shared, local memory file that lives right on your computer. Instead of letting the agent forget why you chose a specific database or why you rejected a certain design pattern, this tool automatically saves those decisions into a tiny, readable file. The coolest part is that it works across different tools, so if you switch from one AI helper to another, your project context doesn't vanish.

📰 https://news.ycombinator.com/item?id=49594008

🆔 @hackernewsgithubprojects

📦 mrajaeim/image-pipes

Image Pipes: Build Vision Pipelines Visually

Image Pipes is the visual editor that turns computer vision coding into a drag-and-drop experience. Instead of writing scripts and guessing results, you connect processing blocks like resizing or blurring on a canvas to see exactly how your image changes in real time. The most surprising part is that once you are happy with your setup, it exports clean, standard Python code. This means you are never locked into the tool; you can take that code and drop it directly into your own projects. It is a great way to experiment with image processing without getting stuck in the code loop.

🆔 @hackernewsgithubprojects

📦 speridlabs/eneas

ENEAS: The AI That Tracks Objects With Words

ENEAS is the video segmentation tool that finds objects using plain English. It solves the headache of manually marking every frame by letting you describe what you see, like the person in the red jacket. The system then tracks that specific item across the entire clip, giving you a clean black and white mask for every single picture. This is great for anyone who needs precise object tracking without drawing boxes by hand. It turns a tedious manual task into a simple text command, saving you hours of work.

🆔 @hackernewsgithubprojects

📦 jdcodes1/facebook-marketplace-mcp

Facebook Marketplace MCP

Turn your AI assistant into a personal shopping scout by letting it search Facebook Marketplace directly. This tool bypasses slow browser automation by reading your existing Chrome login cookies on your Mac and speaking Facebook’s internal data protocol, just like the website does. It’s essentially a reverse-engineered shortcut that lets your AI find specific items, check prices, and monitor new listings in seconds rather than minutes. While it does require a Mac and carries the usual risks of automating a social platform, the speed is surprisingly snappy. If you want your AI to hunt for deals while you sleep, this is the tool to watch.

📰 https://news.ycombinator.com/item?id=47712705

🆔 @hackernewsgithubprojects

📦 step-out/motion-omni

Motion-Omni: The AI That Moves While It Speaks

Motion-Omni is the dialogue model that finally moves its whole body while it talks. Instead of generating speech first and then guessing how to animate a character afterwards, this system predicts facial expressions, hand gestures, and limb movements in the exact same step that it produces the voice. This means a single computational pass creates a fully embodied response, making the character feel much more alive and synchronized than previous approaches. It represents a significant shift towards more natural human-computer interaction by linking language directly to physical movement.

🆔 @hackernewsgithubprojects

📦 liyue-aigc/seedance-2-5-video-director

Seedance 2.5 Video Director: The AI Director's Playbook

Seedance 2.5 Video Director is the prompt toolkit that turns vague video ideas into precise, copy-ready instructions for AI video tools. Instead of guessing what an AI understands, you feed it your raw concept, reference images, or even a bad previous prompt, and it acts like a professional director to organize the chaos. It handles the tricky stuff, like locking a character's face so it doesn't change between shots or ensuring a seamless transition between two clips. It doesn't generate the video itself, but it builds the exact text you need to get the result you want, saving you from endless trial and error.

🆔 @hackernewsgithubprojects

📦 everettfish/holo-card-studio

Holo Card Studio

Describe a character and Holo Card Studio builds a fully interactive three-dimensional collectible card for you. The system generates layered digital art, constructs a physical model in Blender, and exports it into a live web page where the holographic foil actually reacts to your screen angle. This turns a simple text prompt into a dynamic digital object you can spin, flip, and adjust in real time, preserving your specific style and typography. It is a clever way to bridge static design tools with live web technology. Watch the light bend as the card moves. That is the magic you get to keep.

🆔 @hackernewsgithubprojects

📦 layer6ai-labs/cfms

Causal Foundation Models: No Training Needed?

Predict causal effects on new data without training a single new model. This is the coolest part. Causal Foundation Models let you estimate how a treatment affects individuals by simply showing the network a small example dataset. It uses in-context learning to figure out the causal relationship on the fly. You skip the expensive retraining phase entirely. Instead, you just load your data and get instant predictions. This repository gives you the quickest path to doing exactly that. It includes a quickstart notebook and a sandbox to test three different models side by side. You can even run benchmarks against traditional methods to see how they stack up.

🆔 @hackernewsgithubprojects

📦 superxslam/tartanimu

TartanIMU: One Model for All Robots

TartanIMU is the single model that tracks any robot or human using only an inertial sensor. Usually, tracking algorithms need to be built separately for each type of movement, like a car or a dog. This project changes that by training one network on over one hundred hours of data from cars, quadrupeds, drones, and humans. It learns to predict three-dimensional velocity just from acceleration and rotation data. This means one set of weights can handle all these different bodies without needing specific settings for each. If you work on robotics state estimation, this is the universal positioning tool you have been waiting for.

🆔 @hackernewsgithubprojects