📦 realfakenerd/openskyrim

OpenSkyrim: Skyrim Without Loading Screens

OpenSkyrim is building a modern, open-source engine for Skyrim that completely eliminates loading screens. By converting old game files into modern formats like glTF and KTX2, the engine streams assets in real-time, letting you walk from a dungeon straight into the city without ever waiting. This project, inspired by OpenMW, uses Rust and the Bevy engine to handle everything from physics to a new scripting system, aiming for smooth, high-performance gameplay on both desktop and mobile. It’s a massive technical undertaking to bring a classic game into the modern era with zero interruptions, and it’s a fascinating look at how engine reimplementation works.

📰 https://news.ycombinator.com/item?id=49327662

🆔 @hackernewsgithubprojects

📦 danielle34/cavewoman

Cavewoman: Why Talking Like a Caveman Costs You More

CAVEWOMAN is a benchmark that reveals why talking to AIs like a caveman actually costs you more money. The project strips language down to bare nouns and numbers across nine different models to see what breaks. Surprisingly, compressing your input prompts makes models write longer, more expensive replies to compensate for the missing context. On the other hand, forcing the AI to speak in short, blunt sentences genuinely saves you up to three times the money on API fees. It is a clever, data-driven way to find out exactly where your token budget is leaking. Stop guessing which style is efficient and just check the numbers.

🆔 @hackernewsgithubprojects

📦 aigeeksgroup/anydepth

AnyDepth: The Tiny Depth Engine

AnyDepth is the depth estimation framework that shrinks the heavy lifting part by eighty-five percent without losing the accuracy. If your computer vision project needs to figure out how far objects are from the camera, this tool plugs right in and does the job with a fraction of the brain power. It uses a compact decoder that cuts down the mathematical work, making it fast enough to run smoothly on tiny, low-power hardware like a Jetson card. You get professional grade depth maps while keeping your system light and efficient. It is the smart, lightweight way to add spatial awareness to your AI models.

🆔 @hackernewsgithubprojects

📦 olorehq/olore

How olore stops your AI from making things up

Olore is the local documentation tool that stops your AI coding agents from making stuff up. It works like a package manager, but instead of code, it installs full, offline-ready reference books for tools like Prisma or Next JS directly onto your machine. This means your assistant actually reads real, verified documentation instead of guessing, so you get accurate code without needing any cloud infrastructure. Once you install a pack, it links to all your different AI agents at once, keeping everything consistent and private. It is a simple, local first way to give your bots the knowledge they need to get it right the first time.

🆔 @hackernewsgithubprojects

📦 kubeswift-io/kubeswift

KubeSwift: Run VMs Like Kubernetes Pods

KubeSwift is the Kubernetes-native tool that finally lets you run full virtual machines with the same ease as deploying a simple container. It bridges the gap between lightweight microVMs and heavy traditional virtualization by treating every VM like a standard workload you manage with kubectl. The coolest part is the warm-pool sandbox, which keeps pre-booted, isolated VMs ready to go so you can spin up secure environments for untrusted code in fractions of a second. This solves the pain of needing hardware isolation without the slow, clunky setup of a traditional virtualization stack.

📰 https://news.ycombinator.com/item?id=48834952

🆔 @hackernewsgithubprojects

📦 pcm720/dvrpwned

The PSX DVRP Custom Firmware Breakthrough

The PlayStation X finally accepts any hard drive. This new project, DVRPwned, delivers the first working custom firmware for the PSX DVRP. For years, this console locked you into expensive, original Sony hard drives because of a strict security check. This tool patches the internal firmware to bypass that restriction, letting you use almost any standard hard drive instead. The clever part is that it tricks the drive controller into accepting a generic identity, removing the need for expensive proprietary hardware. It is a massive win for retro gaming enthusiasts who want to save their old consoles from failing original disks.

📰 https://news.ycombinator.com/item?id=49328893

🆔 @hackernewsgithubprojects

📦 strelov1/freehire

Freehire: The Job Search Engine That Actually Works

Freehire is the open-source job search engine that cuts through the clutter of modern recruiting. Most job boards are just noisy lists where you spend hours scrolling past stale postings and duplicate entries. Freehire solves this by ingesting data from dozens of different sources, including niche boards and direct company pages, and cleaning it all up in one place. It uses smart background processes to merge similar jobs, flag dead links, and even extract key details like seniority and skills so you can filter for exactly what you need. It turns a chaotic hunt into a clear, organized pipeline.

🆔 @hackernewsgithubprojects

📦 elder-plinius/fable-showcase

One Prompt, 57 Demos

The Fable Showcase is the creative web gallery that proves one prompt can build an entire universe. It starts with a simple challenge to an AI system, asking it to spend its weekly credit limit on generating complex, interactive art. The result is fifty-seven self-contained demos, from simulated fish schools to generative tarot cards, all built in a single day. What makes this genuinely surprising is the engineering behind the curtain. Every single project is just one HTML file with zero external dependencies, meaning you can open any of them directly in a browser without needing a server or internet connection.

📰 https://news.ycombinator.com/item?id=49046110

🆔 @hackernewsgithubprojects

📦 dataelement/dsh-desktop

DSH Desktop: Run AI Agents Like a Native App

Turn complex AI agent setups into a simple desktop experience with DSH Desktop. This application wraps DeepSeek Harness, a powerful local AI runtime, into a native app that handles all the background technical work for you. Instead of manually managing command lines or network ports, you simply open the app, and it automatically starts a secure local instance. The project solves the friction of deploying AI agents by keeping your data safe on your own computer while providing a familiar interface.

🆔 @hackernewsgithubprojects

📦 seitanism/comfyui-h3-motion-context-multiref

The Secret to Endless AI Video

You can now generate twenty continuous clips of a music video where the performer’s face and the original song track stay perfectly locked together without any digital jumps. This ComfyUI node pack, called ComfyUI H3 Motion Context MultiRef, solves the headache of making long AI videos by stitching multiple generations together while keeping the audio and visuals consistent. It uses a clever masking technique to protect the existing song and face, ensuring the AI only generates the new movements on top of your original track. This means you can create extended scenes or full songs without the AI changing the singer’s identity or messing up the beat.

🆔 @hackernewsgithubprojects

📦 weiboai/vibethinker

The Tiny AI That Beats Giants

VibeThinker is the tiny reasoning model that punches way above its weight class. Weibo AI built this one point five billion parameter system to prove that small doesn’t mean dumb. Instead of just throwing more compute at the problem, they used a clever training trick to make the model generate diverse solutions first, then pick the best ones. The result is wild. This small model outperforms DeepSeek R one, a system that is over four hundred times larger, on tough math benchmarks. And here is the kicker. They trained it for under eight thousand dollars, a fraction of what bigger rivals spend.

🆔 @hackernewsgithubprojects

📦 caviraoss/akashic

Akashic: Your Own Live Intelligence Map

Connect aircraft, satellites, and global weather to a single live map without touching a single API key. Akashic solves the headache of fragmented data tools by letting you self-host a powerful intelligence workspace that runs on public feeds alone. The standout feature is its ability to layer live air traffic, seismic events, and strategic infrastructure onto one interactive globe for real-time correlation. You don't just watch a dashboard; you actively explore connections between disparate data points directly on the map. It is built for privacy and simplicity, requiring no external accounts or complex environment setups.

📰 https://news.ycombinator.com/item?id=48966939

🆔 @hackernewsgithubprojects

📦 crabbuild/prolly

Prolly: The Map That Never Loses Data

Prolly gives your app a superpower: it creates snapshots of your data that are completely safe to copy, edit, and delete without ever changing the original. Think of it as a digital time machine for your key-value data. Normally, if you want to try a change without breaking what's there, you have to carefully copy everything or risk losing track of where you started. Prolly uses a clever structure called a prolly tree to make those copies nearly free. You can branch off your main data, make tons of changes, and merge them back in later.

📰 https://news.ycombinator.com/item?id=49325458

🆔 @hackernewsgithubprojects

📦 ccch1mneyyy/working-activity

Make Your AI Agent Talk Back With working-activity

working-activity is the status-line extension that turns your silent coding agent into a chatty companion. Instead of staring at a boring spinning wheel while your AI thinks, you get real-time updates on exactly what it is doing, wrapped in playful, human-like text. It tracks actual tool progress, shows you when the model is stuck, and even drops in rare, colorful visual easter eggs to keep things fun. Whether you are using the Pi CLI or DeepSeek Harness, this project makes the wait for your code to compile feel alive instead of empty.

🆔 @hackernewsgithubprojects

📦 kamalnrf/claude-plugins

Claude Plugins: The App Store for AI Coding Agents

Over sixty thousand distinct AI skills are currently scattered across public repositories, making it nearly impossible for developers to find the right tools for their specific coding tasks. The project known as Claude Plugins solves this fragmentation by acting as a unified registry and command line interface that automatically indexes these resources from GitHub. It allows you to search, install, and manage extensions for major AI coding assistants like Claude Code, Cursor, and Windsurf with a single simple command. The system even includes a special meta skill that lets your AI agent autonomously find and install new capabilities on its own before starting a task.

🆔 @hackernewsgithubprojects

📦 cyqawa/yunx

YunX: The Android App That Cracks Cloud Share Links

Paste a cloud share link and watch it transform into a direct download. YunX is an Android app built to bypass the annoying hurdles of major Chinese cloud drive services like Baidu and Quark. Instead of forcing you to install bloatware just to grab a file, it parses the share URL, extracts the real file location, and lets you download it directly on your device. The coolest part? It uses a built-in browser to handle the login for you. You just sign in once inside the app, and it automatically captures the necessary cookies to keep you connected without any complicated coding.

🆔 @hackernewsgithubprojects

📦 brenpoly/be-more-agent

Be More Agent: The Offline AI Robot for Your Pi

Be More Agent is the offline AI companion that turns a Raspberry Pi into a fully private, conversational robot. Instead of sending your voice data to the cloud, it listens, thinks, and speaks entirely on your own hardware, keeping every interaction local and secure. The project stands out because it acts as a blank canvas, letting you swap in custom face animations and sound effects to create a unique character with a personality of your own. It handles speech recognition, reasoning, and even visual perception without needing an internet connection, making it a fascinating tool for building a personal assistant that respects your privacy.

🆔 @hackernewsgithubprojects

📦 myyangyunfan/dsh_desktop

Double-Click DeepSeek: No Setup Needed

Run a full AI agent on your desktop without touching a single terminal. DSH Desktop bundles the entire DeepSeek Harness runtime and Node environment into one neat package, meaning you can double-click a file and be chatting in seconds. It skips the usual headache of installing dependencies or configuring ports, turning a complex command-line tool into a simple, native window that even lives in your system tray. The best part is how it handles your workflow; it includes a tiny whale pet that sits on your desktop to keep you company while your tasks run.

🆔 @hackernewsgithubprojects

📦 webpro-nl/unbash

unbash: The Fast Bash Parser

Inspecting bash scripts without running them sounds risky, but unbash changes that. This TypeScript tool parses shell commands instantly, giving you a clear, typed map of the structure without executing a single line. It is incredibly fast, handling massive files at speeds that make older parsers look like they are stuck in the past. The real magic is how it handles nested commands, letting you see exactly where variables and subshells live so you can audit your code safely. It even recovers from broken input, so your tools never crash on bad data.

📰 https://news.ycombinator.com/item?id=47333545

🆔 @hackernewsgithubprojects

📦 gluonfield/jaz

Jaz: Your AI Never Sleeps

Jaz lets your coding agents run loops overnight on your own machine. It is a personal AI setup that connects to Codex, Claude, and other tools so they can keep working while you sleep. The software uses a client and server split, meaning the heavy lifting happens on a backend you control. You can plug in your existing subscriptions, so you are not paying extra for a new platform. It adds memory that sticks around and lets you push code directly to git. The result is a powerful, always on assistant that lives on hardware you own, not in some closed cloud service.

🆔 @hackernewsgithubprojects