📦 alchaincyf/3d-vibe-coding-handbook

3D Vibe Coding Handbook

3D Vibe Coding Handbook is the interactive companion that turns abstract coding concepts into tangible, spinning mechanical reality. It serves the official HTML version of the handbook alongside a gallery of working web demos, such as a fully functional four-stroke engine. The standout feature here is the engine simulation, which you can physically disassemble, rotate, and even reassemble in a strict, real-world order to truly understand how the parts interact. It is not just a static diagram, but a playable laboratory for spatial reasoning. Grab this repository to see exactly how code builds complex, interactive 3D systems that actually work.

🆔 @hackernewsgithubprojects

📦 ahernresearch/grava

How GRAVA Makes Self-Driving Cars Think

Self-driving cars can now show their work by linking what they see to the exact steering move they make. This new framework, called GRAVA, solves a major mystery in autonomous vehicles: why does the car brake or turn? Instead of just outputting a command, the system builds a clear connection between visible objects and the final driving action. It uses a single model to describe the scene and then directly generates the physical path the car will take. This transparency is critical for trust and safety. As the code and data roll out, we finally have a way to see the logic behind the wheel.

🆔 @hackernewsgithubprojects

📦 ziqin-h/3dway

3DWay: Turning 2D Images into 3D Robot Moves

A vision-language model now plans robot movements by predicting 2D waypoints in two camera views, then uses simple geometry to reconstruct a precise 3D path. This project, called 3DWay, solves the tricky problem of getting a language-based AI to control a physical robot arm without noisy depth sensors. By letting the AI reason in its native image space and letting math handle the 3D conversion, it creates reliable trajectories that work on real robots. It’s a clever way to bridge the gap between digital reasoning and physical action, making robot control more general and less dependent on specialized hardware.

🆔 @hackernewsgithubprojects

📦 omni-interaction-gander/omni-interaction-agent

Gander: Talk While It Works

Delegate a heavy task and keep talking, interrupting, or sharing your screen while the agent keeps working. Gander is an open-source voice assistant that doesn't pause to listen; it hears, sees, and thinks all at once. Most bots freeze while they process, but Gander lets you redirect it mid-sentence or show it your screen without stopping the job. It runs on a split brain, with one part handling real-time chat and another crunching the long, hard tasks in the background. You can just chat naturally, like a friend, while it figures out the details. It’s a cool look at how future assistants might actually behave.

🆔 @hackernewsgithubprojects

📦 vaderchen/yourdesk

YourDesk: The AI Remote Desktop

YourDesk lets an AI agent physically drive your computer. Built in Go for Mac and Windows, this cross-platform tool handles the heavy lifting of video streaming and screen input so your software assistant can click, type, and navigate just like a human. It is not a basic screen viewer; it is a direct control channel for your favorite AI models. Because the code is open and free to use for internal work, developers can build on it without licensing headaches. If you are building autonomous tools, this is the bridge between the language model and the actual machine. Give your AI hands, not just a brain.

🆔 @hackernewsgithubprojects

📦 keesan12/martin-loop

MartinLoop: The Safety Net for AI Coding Agents

MartinLoop stops AI coding agents from spiraling out of control. When an AI tries to fix code, it can accidentally run endless loops that burn through your money without finishing the job. This project wraps the agent in a safety net that sets a hard limit on spending and checks the work against real tests before calling it done. If the agent fails, the system rolls back the changes so your code stays safe. You get a clear receipt showing exactly what happened and why it stopped. It turns a risky guess into a proven, reviewable result.

📰 https://news.ycombinator.com/item?id=48098126

🆔 @hackernewsgithubprojects

📦 huang-sh/pix

PiX: Chat With Branches

PiX turns your AI chat into a growing map, not a single line. Instead of getting stuck in a messy thread, you can branch off any moment to try a new idea, leaving the original path exactly as it is. If you hit a dead end, you just pick up from that point and try another route. The coolest part is seeing three different versions of your conversation side-by-side, letting you compare answers without losing your place. It feels like having parallel universes for your thoughts, all in one window. If you have ever felt your AI context getting too tangled, this is the fix. Stop piling on; start branching.

🆔 @hackernewsgithubprojects

📦 wilbowes/echomuse

EchoMuse: The $10 Local Voice Assistant

Your ten dollar Echo Dot is secretly a powerful local server. EchoMuse is an open source project that rips out the Amazon cloud dependency and turns that cheap hardware into a fully private, offline voice satellite for Home Assistant. You plug it in, say a custom wake word, and it talks directly to your local AI without sending a single word to a server. It handles music, lights, and even uses the built in ambient light sensor that Amazon never exposed. This is the definitive guide to taking that forgotten device from a digital spy and making it a genuinely useful, local part of your smart home setup.

📰 https://news.ycombinator.com/item?id=49722846

🆔 @hackernewsgithubprojects

📦 penn-rail/marc-v1

MARC: The AI Framework That Thinks Step-by-Step

MARC is the multi-agent framework that finally breaks down complex reasoning into visible, manageable steps. Instead of asking a single artificial intelligence to guess an answer in one go, this system splits the task among a team of specialized digital workers. Each one handles a specific part of the puzzle and passes their findings clearly to the next person in line. This transparency makes it easy to see exactly how the final result was reached. It is a brilliant way to build systems that are easier to check and improve. You get a clear view of every logical leap the machine takes.

🆔 @hackernewsgithubprojects

📦 yjz211/vivid-figures-skill

Vivid Figures: AI-Generated Scientific Charts

One hundred and eight distinct recipes allow an AI assistant to transform your raw data into publication-ready scientific figures. This repository solves the tedious problem of manually coding complex visualizations by providing a complete library of chart templates ranging from standard bar graphs to intricate heatmaps. It automatically selects the right visualization for your specific data structure and applies one of seven curated color palettes to ensure the final image looks professional and consistent. The system delivers both the rendered image and the underlying source code, giving you full control. It is a practical tool for researchers who need clear, beautiful data visualizations without spending hours on design details.

🆔 @hackernewsgithubprojects

📦 saragordic/window-sweaters

Window Sweaters: Knitted Borders for Your Mac

Your Mac windows are literally wearing knitted sweaters. This tiny menu bar app draws animated, pixel-perfect yarn borders around every active window, turning your flat screen into a cozy, tactile workspace. It isn't just a generic frame; each application gets its own hand-picked colorway and pattern, so your code editor looks completely different from your music player. It tracks your focused window in real time, keeping the borders perfectly aligned even as you move or resize things. It is a purely cosmetic tool that proves you can inject serious personality into your operating system.

📰 https://news.ycombinator.com/item?id=49717555

🆔 @hackernewsgithubprojects

📦 haolpku/worksurface-bench

How to Tell if Your AI Agent Knows Where to Look

WorkSurface-Bench is the benchmark that finally checks if your AI agent knows which tools to grab. Most agents just guess, but this project tests if they can actually route to the right source. It gives them documents, tables, and maps to choose from. Then it scores if they picked the right one. It separates smart routing from lucky guessing. That is the real test of enterprise intelligence. So you know exactly where your agent stands.

🆔 @hackernewsgithubprojects

📦 xhluca/session-migrate

Session Migrate: Move Your Coding History

Session Migrate lets you lift a live coding conversation out of one AI tool and drop it straight into another, keeping the full context intact. Instead of pasting text or starting from scratch, this tool reads the native session file from an agent like Claude Code or Codex and writes a new, resumable session into a completely different harness, like Pi or Copilot. It handles the messy translation of messages and tool results so the new agent knows exactly where you left off. It is not just an export feature; it creates a fresh, native session that the target application can pick up and continue as if nothing happened.

📰 https://news.ycombinator.com/item?id=49425594

🆔 @hackernewsgithubprojects

📦 skillsynchq/txcript

Continue Claude Chats in Codex with txcript

Pick up a coding conversation where you left off, even if you switched tools. txcript lets you export a full session from one AI assistant, including every message, tool call, and reasoning step, and import it directly into another. This solves the annoying problem where your AI coding agent forgets everything when you change platforms. Instead of starting over, you carry your entire work history across different coding environments like Claude Code and Codex. It acts as a universal translator for your AI chats. If you are moving projects between tools, you do not have to re-explain your context.

📰 https://news.ycombinator.com/item?id=49719696

🆔 @hackernewsgithubprojects

📦 tencentcloud/octop

Octop: Your Private AI Team

Build a private AI assistant that actually runs on your own hardware. Octop is a self-hosted system that lets multiple people use separate, specialized AI agents on a single server. It solves the problem of generic chatbots that lack context or privacy, giving you full control over who can access what. The coolest part is that it supports multi-agent setups, meaning you can have different digital workers handle different tasks for different team members simultaneously without them getting confused. It is essentially a private, customizable office where your data never leaves your walls. If you want total ownership over your AI experience, this is your playground.

🆔 @hackernewsgithubprojects

📦 meituan-longcat/longcat-video

LongCat Video: The Open Source Video Generator

LongCat Video is the open-source video model that finally keeps long generations stable without the usual quality drop. It solves the headache of stitching together short clips by natively creating minutes-long footage where colors and details stay consistent from start to finish. Developed by Meituan, this thirteen billion parameter tool handles text-to-video, image-to-video, and even audio-driven character animation all in one place. You can run it on your own hardware to generate high-definition footage quickly using a smart generation strategy that balances speed and quality. It is a huge step for anyone wanting to build their own cinematic content without relying on closed, expensive commercial services.

🆔 @hackernewsgithubprojects

📦 vercel/eve

Eve: Code Agents Using Plain Files

Build AI agents using nothing but plain text files. Vercel’s Eve framework treats your folder structure as the source of truth, meaning your agent’s instructions, tools, and schedules all live right in front of you. This solves the black box problem because you can actually read, edit, and inspect exactly what your AI is doing. It’s a filesystem first approach that makes debugging and customization far more intuitive than hidden configuration files. By keeping everything visible and standard, you get a project you truly own. Next time you build an agent, try dropping a simple markdown file into a folder and see how much control you gain over the behavior.

🆔 @hackernewsgithubprojects

📦 fluxvla/fluxvla

FluxVLA: One Platform for Real Robot AI

Deploying robot AI usually means wrestling with mismatched codebases and slow inference pipelines. FluxVLA solves this by providing a single, unified platform that handles everything from data processing to running models on physical hardware. The standout feature is its ability to run heavy vision language models directly on compact edge devices like the Jetson Orin. This means you don't need expensive cloud servers to make a robot think and act in real-time. By standardizing the interface, it removes the technical friction that usually slows down research.

📰 https://news.ycombinator.com/item?id=49724566

🆔 @hackernewsgithubprojects

📦 nhtera/oximux

OxiMux: Parallel Coding Agents for Mac

OxiMux lets you run multiple AI coding assistants at the same time on a single Mac. The most surprising part is that it gives every agent its own private, isolated copy of your code project. This means Claude and Codex can work on different features simultaneously without ever breaking each other's work. The tool handles all the messy setup automatically, so you just focus on the code. It turns your computer into a multi-agent cockpit where you can review changes side by side. If you want to get more done without the chaos of manual branching, this is the tool to watch.

🆔 @hackernewsgithubprojects

📦 jedisct1/turbocrypt

TurboCrypt: Keep Private Files in Public Git Repos

TurboCrypt lets you commit private notes to a public Git repository without anyone seeing the actual content. This command line tool encrypts files so fast that it handles entire directory trees in parallel, keeping your secrets safe even when the code itself is public to the world. It solves the common problem of needing to share encrypted files with teammates while keeping them out of reach from everyone else. The software uses modern cryptography to secure your data and works on almost any operating system. You can also mount these encrypted folders to work with your normal apps.

📰 https://news.ycombinator.com/item?id=49725409

🆔 @hackernewsgithubprojects

📦 maelic/relateanything

RelateAnything: Any Relation, Any Image

RelateAnything identifies relationships between objects in an image using words you provide at the moment of analysis, rather than a fixed list trained into the system. This small, fast model processes visual data in real time to find connections like riding or standing next to, even for concepts it has never seen before. It solves the problem of rigid software that fails when you need a specific, unusual connection. Because it is tiny and lightweight, it runs instantly on standard hardware, making it easy to build applications that understand complex interactions in video streams. You can define exactly what relationships matter to your project without retraining the whole system.

📰 https://news.ycombinator.com/item?id=49713313

🆔 @hackernewsgithubprojects