📦 alphanu1/daytona-arcade-recomp

Daytona USA: The Arcade Game Rebuilt From Scratch

Daytona USA is the arcade racer that ditched emulation entirely to run the original hardware code directly on your computer. Instead of simulating old chips, developers took the game's processor instructions and translated them into modern, portable code, while the graphics engine builds every frame from scratch using your own video card. This means the car, the track, and the sound board all run natively on your desktop, delivering a speed and accuracy that traditional emulators simply cannot match. It is a rare look under the hood of a classic machine, proving that a nineteen-nineties racing game can still feel brand new when rebuilt from the ground up.

🆔 @hackernewsgithubprojects

📦 wenyiwu0111/rsigame

RSIGame: The Game That Builds Itself

A computer program can now build a playable video game, find its own bugs, fix them, and keep getting better on its own until it hits a limit. This is how RSIGame works. It uses AI to play the game, watch for glitches, and repair the code in a loop. The coolest part? It doesn't just guess what’s wrong. It records evidence, like screenshots and logs, to pinpoint exactly where the game breaks. Then, it fixes that specific issue, checks the whole game again to make sure nothing else broke, and saves the best version.

🆔 @hackernewsgithubprojects

📦 quimmedes/cafe-llama.cpp

Run Massive AI Models on Your Laptop

Run massive AI models on your laptop without expensive hardware. Cafe Llama C P P is an experimental engine that takes standard open source models and lets them breathe life into your computer. It solves the problem of memory limits by intelligently moving heavy parts of the model to your regular system memory or even your hard drive when your graphics card runs out of room. This means you can run complex reasoning tasks locally with surprising speed. It is a powerful way to keep your data private while still accessing advanced intelligence. Give it a try if you want to build AI apps that run anywhere.

🆔 @hackernewsgithubprojects

📦 chinesezjc/cordisclaw

CordisClaw: Plugins That Prove Themselves

CordisClaw builds a Rust runtime that treats plugins like strict legal contracts rather than loose code snippets. The most surprising thing it does is verify every plugin against machine-readable documentation before it ever runs. This means the system knows exactly what each piece of code can do and who it is allowed to talk to, preventing silent failures and hidden dependencies. It is a deliberate move away from messy dynamic loading toward a fully auditable, fail-fast architecture. If you are frustrated by plugin systems that break for unclear reasons, this project shows how making rules explicit turns chaos into clear, debugable logic.

🆔 @hackernewsgithubprojects

📦 romancypherpunk/12axes

12 Axes: A Political Quiz That Refuses to Simplify

Place yourself on a political map that refuses to flatten your complex beliefs into a single left or right label. The project called Twelve Axes measures twelve independent dimensions separately, from economic control to social morality, so you can see exactly where your views clash or align without averaging them out. By comparing your specific combination against hundreds of ideologies, historical regimes, and political figures, it reveals the unique tensions in your worldview that standard tests hide. It is a fascinating way to understand exactly where you stand relative to the world, offering a detailed portrait of your political identity in plain English.

🆔 @hackernewsgithubprojects

📦 continuum-ai-corp/orcapromptvault

OrcaPromptVault: The Hidden Rules of AI

Peek behind the curtain of your favorite AI tools. OrcaPromptVault preserves the exact instructions that power chatbots like ChatGPT, Claude, and Grok. It solves the mystery of why an AI behaves a certain way by archiving the specific text sent before you type a single word. Instead of guessing, you can verify exactly what the machine is told to do. This turns a black box into a clear, readable document. No more folklore about hidden rules, just the facts. Check it out to see how these systems really think.

🆔 @hackernewsgithubprojects

📦 chenxiachan/paperfold

PaperFold: Zoom Into Any Research Paper

PaperFold is the local paper reader that lets you zoom from full text to a single-word topic map. It solves the headache of holding a complex argument in your head by folding sections into concise summaries while keeping your place. As you pinch to zoom, words slide smoothly to their new spots so you always know exactly where you are. It even highlights your notes as you zoom out. If you read academic papers, this tool makes the structure instantly clear without losing the original detail.

🆔 @hackernewsgithubprojects

📦 coral79/pami-code

PAMI: Text to Human-Object Interaction

Turn text descriptions into realistic full-body human-object interactions. PAMI generates coordinated motion where people pick up, lift, or move objects without drifting or clipping through them. The key innovation is part-anchored motion, which tracks object position relative to specific body parts rather than treating them as separate entities. This approach solves common problems like missed contacts and penetration that plague previous methods. By using a coarse-to-fine generation hierarchy, PAMI first creates a general interaction from text, then refines it with detailed surface sensing for accurate contact geometry. The result is more faithful and physically plausible human-object interactions than existing approaches.

🆔 @hackernewsgithubprojects

📦 josevalim/lynx

Lynx: Proving Your Elixir Code Is Correct

Lynx is the proof engine that guarantees your Elixir code actually works as promised. Instead of just running tests to see if the output looks right, it uses a formal logic tool to mathematically verify that your functions follow specific rules you define. You write a simple statement about what your code should do, and the system automatically translates your Elixir into a language the proof checker understands. It then rigorously checks every possible input to confirm your code never breaks that rule.

🆔 @hackernewsgithubprojects

📦 pepe-agent/pepe

Pepe: The AI Agent That Actually Finishes Jobs

Point an AI at a specific outcome and let it work until a separate reviewer confirms the job is done. That is the core of Pepe, a self-contained runtime built in Elixir that connects to any model you choose. Unlike standard chatbots that just answer questions, this system runs a loop where an agent executes tasks and an independent judge checks the result against your exact criteria. If the output fails, it retries automatically until it meets the standard. It is not just a chat window, but a full engine with a web dashboard and phone integrations. You define the goal, and the system handles the work, verification, and retries.

🆔 @hackernewsgithubprojects

📦 wobushannes/drehbuch--referenzanker-generator

How this tool fixes AI video prompts

This open-source studio crushes your scattered bullet points and character photos into one unbroken line of text for AI video generators. Creators lose hours to errors when prompts contain line breaks, which breaks the code and ruins the visual output. This app solves that by using a local vision model to look at your reference images and extract the exact details needed. It then merges everything into a strict, timecoded string that runs without interruption. You get a clean, professional prompt that works every time. It is a simple way to make your AI video production more reliable and professional.

🆔 @hackernewsgithubprojects

📦 wurli/servery.nvim

servery.nvim: Switch Neovim Sessions Instantly

servery.nvim is the Neovim plugin that finally makes switching between editor sessions feel as fast as searching your files. It solves the frustration of juggling multiple projects by letting you jump between active editor instances using the fuzzy finder you already love. Instead of wrestling with terminal multiplexers or complex scripts, you just open your usual picker, see your open sessions listed right next to your current one, and press enter to teleport. The sessions keep running in the background even if you detach, so your work stays alive. If you edit multiple projects at once, this cuts the clutter down to a single, lightning fast keystroke.

🆔 @hackernewsgithubprojects

📦 graze-social/cocore

cocore: The AI Co-op

The compute co-op that puts the ledger in your hands. cocore turns the spare processing power hiding in your closet into a shared resource for everyone else. Instead of renting from a giant cloud company, you plug your own machine into a network where people lend their hardware to run AI tasks for each other. The coolest part is the transparency. Every single job creates a signed receipt that lives on the blockchain. This means anyone can verify exactly what work was done and who did it without trusting any central server. It is a genuine experiment in sharing.

🆔 @hackernewsgithubprojects

📦 comm4nd0/conflict-map

conflict map

Track the world's active conflicts on a living globe that runs entirely on your own computer. Conflict map pulls live news from major outlets and global event data, then uses a local artificial intelligence model to identify exactly who is fighting, who is supporting whom, and what the latest developments are. The most striking feature is how it visualizes reported attacks as animated projectiles flying across a detailed 3D globe, giving you an immediate, visual sense of where violence is occurring right now. Because it processes everything locally, your data stays private, and you get a clear, fact-based picture of current events without relying on a distant cloud server.

🆔 @hackernewsgithubprojects

📦 yuutatsubasa/free-riders-recompiled

Free Riders Recompiled

Sonic Free Riders is no longer trapped on the Xbox Three Sixty or bound to the Kinect. This is achieved by free-riders-recompiled, an unofficial port that statically recompiles the game to run natively on Windows, Linux, and Android. By translating the original console code and shaders into modern computer instructions, it completely bypasses traditional emulation. The project fully emulates the original motion controls, letting you steer with a standard controller, touch screen, or even a basic PC webcam that estimates your physical body movements. It is an incredibly clever way to preserve and play a forgotten classic on modern hardware.

🆔 @hackernewsgithubprojects

📦 metaevo-ai/harness-zero

Harness-Zero

Banish bloated software wrappers and bake those clever agent behaviors directly into your AI model's weights during training. Heavy frameworks like external databases, memory tools, and custom middleware make AI agents perform better, but they slow everything down at deployment. This project solves that by letting a smart teacher agent watch your model work, intercept its mistakes, and rewrite its actions using the best possible strategies. By training your model on these optimized paths, it actually internalizes the advanced behavior. When you deploy, you can ditch the clunky extra scaffolding entirely, leaving you with a lightweight, lightning-fast model that still gets elite results.

🆔 @hackernewsgithubprojects

📦 zcanann/sfa-decomp

sfa-decomp: Rebuilding Star Fox Adventures

Sfa decomp is the reverse engineering project that reconstructs the lost source code of Star Fox Adventures for the GameCube. This massive effort tackles the challenge of reading raw machine code to recover the original human readable logic that developers wrote years ago. The team documents every function and data structure to build a working replica of the game engine from scratch. It proves how much hidden knowledge exists inside old software if you look closely enough. If you have ever wondered how games are built, this project shows you exactly how the sausage is made.

🆔 @hackernewsgithubprojects

📦 luwudynamics/xgoduck_rl

Teach a Robot to Kick a Ball

Teach a tiny bipedal robot to kick a ball and navigate rough terrain without falling over. That is exactly what XgoDuck RL achieves by creating specialized simulation environments for a specific physical robot design. It solves the problem of training complex movement policies on real hardware by running thousands of virtual trials simultaneously on your graphics processing unit. This is particularly interesting because it bridges the gap between theoretical algorithms and a physical machine you can actually build. You train the brain in the computer, then download the results directly onto the small robot to see it move in the real world.

🆔 @hackernewsgithubprojects

📦 karolynaz/devil-3ds

Diablo on 3DS: Finally Comfortable

Devil-3Ds lets you actually enjoy playing Diablo on your Nintendo 3DS by completely reworking the controls and layout. The previous port forced you to use a desktop-style interface on a tiny handheld, which was frustrating to navigate. This new version fixes that by moving your health, mana, and item belt to the dedicated lower screen, while the top screen handles the main gameplay. You get touch support for your belt slots and a cleaner, more readable interface for managing your inventory and spells. It’s a focused effort to make the classic action RPG feel at home on this specific hardware.

🆔 @hackernewsgithubprojects

📦 skyvanguard/awesome-ai-pentesting

AI Hackers: The Future of Pentesting

An AI agent just found and exploited a real security flaw without a single human typing a command. That is the reality inside a curated collection called awesome ai pentesting. It is not a single program but a living map of the open-source world where artificial intelligence meets offensive security. The list tracks autonomous agents that can scan networks, find weaknesses, and even prove they are real by breaking in themselves. It solves the problem of speed and scale, letting machines handle the tedious reconnaissance so humans can focus on strategy. If you are curious about how code can hunt bugs, this is your starting point.

🆔 @hackernewsgithubprojects