📦 framefilter/keyroost

Keyroost: One App for All Security Keys

Keyroost lets you manage a YubiKey, a SoloKey, or a Token2 fob with one single app, no matter who made it. Most tools only work with one brand, but this works because it speaks the universal language of open standards instead of relying on private software from each maker. It handles everything from setting up fingerprints to generating encryption keys and resetting a stuck device. The coolest part is the factory reset: it wipes every part of the key in the right order, so you can start fresh without guessing.

🆔 @hackernewsgithubprojects

📦 benoitc/erlang_wasm

erlang_wasm: The WebAssembly Runtime Written in Erlang

erlang_wasm is the WebAssembly runtime that runs entirely inside the Erlang virtual machine without any heavy native dependencies. It lets you execute code compiled to WebAssembly directly inside your Erlang applications, which solves the tricky problem of safely running untrusted or multi-language code without leaving your existing system. The really surprising part is how it handles memory, using clever techniques to make it fast enough to compete with runtimes written in C. If you work with Erlang and need to run external code, this gives you a powerful, contained way to do it.

🆔 @hackernewsgithubprojects

📦 magicbug/oscarwatch-tracker

OscarWatch Tracker: Satellite Pass Automation for Ham Radios

OscarWatch Tracker automatically controls your radio frequency and antenna rotator to track moving satellites in real time. This desktop application solves the problem of manually calculating complex Doppler shifts and aiming antennas for brief satellite passes. It calculates the exact frequencies needed for uplink and downlink signals while accounting for the satellite's speed and your specific location. For amateur radio operators, this means less mental math and more focus on making clear contacts. The software supports various radio models and rotator controllers, handling the technical tracking work in the background. You get a reliable way to track these fast-moving objects without constant manual adjustments.

🆔 @hackernewsgithubprojects

📦 pepasjc/snatcher-translated

Snatcher Translated: Full English Patch

Kojima's cyberpunk masterpiece Snatcher never officially left Japan in English, until now. This fan-made patch translates the entire 1996 game for both Sega Saturn and PlayStation, handling every single line of dialogue, menu, and even the voice acting subtitles. The real challenge here is the technical engineering; the developer had to reverse engineer the original code to fit English text into a game built for dense Japanese characters. By redrawing fonts and modifying the memory structure, they ensured no lines are cut off, giving you a complete, uncut experience on real hardware or emulators.

🆔 @hackernewsgithubprojects

📦 zeldawwhdrecomp/zeldawwhdrecomp

Zelda Wind Waker HD Recomp

Zelda Wind Waker HD Recomp is the native macOS port that finally runs the Wii U classic on Apple Silicon without emulation overhead. It translates the original PowerPC code directly into C, completely bypassing the standard Cemu runtime to create a lightweight, high-performance experience. The coolest part is the true sixty frames per second mode, which runs the game logic at full speed instead of just visually smoothing things out, making gameplay feel significantly snappier than the original. Since the repository only contains the runtime tools and requires your own legally obtained disc dump, it is a pure fan project for enthusiasts.

🆔 @hackernewsgithubprojects

📦 yzmblog/dmad

DMAD: 4-Step Video and Audio Generation

Turn a slow, heavy video model into a lightning-fast creator that generates both moving pictures and matching sound in just four quick steps. This project, called DMAD, takes a massive thirty-three billion parameter teacher model and distills its skills into a compact student. The trick is clever: instead of the usual slow, memory-hungry fitting process, it uses a discriminator to teach the student the difference between real and generated data directly. The result is a tool that makes high-definition video clips with synchronized stereo audio without waiting around for dozens of iterations. It’s a neat, efficient way to get professional-looking results in seconds.

🆔 @hackernewsgithubprojects

📦 shulin16/v-rubrics

How V-Rubrics Teaches AI to See Better

Make your AI see what it’s actually looking at, not just what it thinks it sees. V-Rubrics solves the classic problem where visual models confidently describe things that aren’t there. Instead of just checking if the final answer is right, this project breaks down every response into tiny, specific rules. It then uses reinforcement learning to reward the model only when it sticks to those strict visual facts. The coolest part is how it catches the model hallucinating a detail and immediately penalizes that exact mistake. It’s like giving the AI a strict teacher who checks its work line by line.

🆔 @hackernewsgithubprojects

📦 xqlsystems/duckdb-zarr

DuckDB-Zarr: SQL for Scientific Data

DuckDB-Zarr lets you point your favorite SQL queries at massive scientific datasets stored as Zarr files, turning complex climate grids into simple tables. The coolest part is how it handles time: instead of giving you raw, cryptic numbers that require messy manual math, it automatically converts those values into real calendar dates right when you run your query. That means you can finally use standard date filters and functions without the headache, making it easy to analyze weather or ocean data directly from the cloud. If you work with scientific arrays and love the simplicity of SQL, this tool is a must-try.

🆔 @hackernewsgithubprojects

📦 xueshiqiao/qduo

QDuo: Your Text, Transformed Instantly

QDuo turns the simple act of selecting text on your Mac into a command center for instant transformation. When you highlight a word or sentence in any app, a tiny menu appears right at your cursor, ready to translate, summarize, or explain the text without ever leaving your current window. This small macOS tool removes the friction of copying, pasting, and switching tabs, letting you work with the words on your screen directly. It connects to AI models or runs simple text commands to fix grammar or change case, keeping your workflow focused and your hands on the keyboard.

🆔 @hackernewsgithubprojects

📦 supercmohq/supercmo-skills

SuperCMO Skills: The AI Marketing Agency

Hand a single product photo to an AI agent and watch it build a finished, scroll-stopping ad video from scratch. That is the core promise of SuperCMO Skills, an open-source toolkit that turns your coding agent into a full-service marketing studio. Instead of jumping between separate tools for scripting, voiceover, and video rendering, this system handles the entire pipeline in one place. It picks the right AI models for the job, casts consistent AI actors, and stitches every clip into a final video that you can approve step by step. The result is a professional, campaign-ready asset without needing a traditional production team.

📰 https://news.ycombinator.com/item?id=49888905

🆔 @hackernewsgithubprojects

📦 archestra-ai/openappa

OpenAPPA: The Security Gate for AI Agents

OpenAPPA acts as a strict security gate that stops AI agents from leaking sensitive data before any action happens. Instead of relying on unpredictable keyword filters that might miss a leak, this tool tracks exactly where every piece of data comes from and checks it against a fixed set of rules. If the data is too sensitive for the destination, the system blocks it instantly, ensuring that private information never reaches the wrong tool. This deterministic approach means the same input always gets the same result, which is critical for handling serious records in real work.

🆔 @hackernewsgithubprojects

📦 runwisp/runwisp

RunWisp: Your Server's Black Box

A single Go binary now acts as a complete black box for your server's background tasks. RunWisp combines standard scheduling with process supervision, meaning it handles both timed jobs and long-running services without needing Python or external databases. The standout feature is that every execution leaves a permanent record, storing the exit code, duration, and full console output in a built-in database. This means you can finally see exactly why a job failed and what it printed, right from a web dashboard or terminal interface. It replaces the need for separate cron and supervisor tools, giving you total visibility into your server's activity.

🆔 @hackernewsgithubprojects

📦 elripalda/porpoise

Porpoise: GameCube and Wii on PS5

You can play your old GameCube and Wii discs directly on a PS5, and Porpoise does it without needing a separate computer. It wraps the Dolphin emulator into a native app that sits right on your home screen, so your PS5 hardware does the heavy lifting. The library looks like a sleek row of box art tiles, complete with original music and a controller setup that actually maps the DualSense to feel like a Wii remote. It is a polished, standalone experience for your own backups, meaning no weird file managers or command lines, just pick a game and play.

🆔 @hackernewsgithubprojects

📦 rauchg/gdp-ts

GDP-TS: Stop Authorization Bugs

GDP-TS is the TypeScript tool that makes authorization bugs impossible at compile time. It solves the classic problem where a developer forgets a check or mixes up user IDs, causing a security leak. Instead of just checking permissions, you pass a special ghost proof to sensitive functions. If the proof is missing or wrong, the code simply will not compile. It costs almost nothing at runtime and fits into any project. The type checker does the heavy lifting, so you catch mistakes before they ever reach production. It turns a risky manual check into a guaranteed safety net.

🆔 @hackernewsgithubprojects

📦 christianmat/jev-pokemon

Jev Plays Pokémon Red

Jev, the AI decision model from TypeSafe, is playing the entire main story of Pokémon Red with zero cheat codes or scripts. It runs inside a real emulator, reading the console's memory to see the world and then picking the very next button press itself. There is no hidden game knowledge here; it does not know where items are hidden or how to beat trainers without training. It has to figure it all out by walking, battling, and talking, just like you would. The best part is that it costs about a dollar to watch it play for a whole day.

🆔 @hackernewsgithubprojects

📦 andududu/jeview

Jeview: Watch Your Code Think

Jeview lets you watch every decision your code makes in real time. It runs on your own computer and sits between your app and TypeSafe, capturing each request and answer into a local database. The coolest part is the live map: it draws a growing tree of your logic as it happens, so you can actually see how one answer triggers the next. No cloud, no accounts, just a simple browser window showing your data flow. It’s a rare peek inside the black box, perfect for developers who want to understand exactly what their system is deciding and why. Give it a spin to finally see your logic move.

🆔 @hackernewsgithubprojects

📦 shootthesound/comfyui-fizgig-h3-still

comfyui-fizgig-h3-still

You can now extract stunning, crisp high-resolution still images directly from the MiniMax H3 video model without rendering heavy multi-frame clips or dealing with ugly, banded artifacts. A clever custom node extension called comfyui-fizgig-h3-still introduces two dedicated nodes that generate a true one-frame latent and decode it cleanly at virtually any size. By bypassing the default video decoding process, it preserves incredible skin detail and texture while running faster than standard workarounds. This makes it incredibly easy to create ultra-detailed image generation and editing workflows using your existing setup. Grab these nodes to start exporting gorgeous single-frame masterpieces today.

🆔 @hackernewsgithubprojects

📦 akkie76/code-review-skills

The AI Code Review That Doesn't Hallucinate Bugs

Code Review Skills is the instruction set that makes A.I. code reviewers actually stick to the facts. Instead of letting your coding agent guess or invent fake bugs to sound helpful, this tool forces it to provide concrete evidence for every single issue it finds. It completely ignores vague style complaints and only flags real problems with a clear impact. Think of it as a strict quality control layer for your A.I. assistants. If you are using tools like Codex or Claude Code to write software, this is the missing piece that keeps your reviews honest and useful. No more guessing, just verified facts.

🆔 @hackernewsgithubprojects

📦 psrben/visionhope

VisionHOPE: The AI That Learns While Looking

A new vision model called VisionHOPE doesn't just look at images; it learns to look at them in real time. Standard neural networks have fixed rules for processing data, but this one changes its own internal memory and learning rules as it scans a picture. This is called self-referential nested learning. Instead of using the same static filters for every pixel, the model adapts its entire processing strategy on the fly, based on what it has already seen in that specific image. This allows it to handle complex visual tasks like object detection and segmentation with a unique, dynamic approach.

🆔 @hackernewsgithubprojects

📦 lateos-ai/reflex

Reflex: The Cold-Start Inference Engine

Reflex is a Rust inference engine that delivers its first result in under half a second from a completely cold start. Unlike traditional servers that sit idle waiting for work, this tool runs one single task, finishes instantly, and shuts down to stop the meter. It precompiles its graphics math at build time, so there is no delay when the computer first wakes up. This makes it perfect for serverless platforms that charge by the second, where speed is the only thing that matters. It beats heavy competitors like vLLM by over three hundred percent on startup time. You get fast, single-shot answers without paying for waiting rooms.

📰 https://news.ycombinator.com/item?id=49852130

🆔 @hackernewsgithubprojects

📦 leaplabthu/simple-wam

Simple-WAM: The Robot That Dreams Before It Moves

Simple-WAM is a robot learning model that actually predicts the future video of what it is about to do before it moves a single joint. That sounds like sci-fi, but it is exactly how this system learns to generalize to new tasks it has never seen. Instead of just memorizing button presses, the model builds a mental simulation of the outcome, letting it adapt on the fly. This research code releases the exact training scripts and evaluation tools used in the paper, so you can see exactly how these world action models work under the hood.

🆔 @hackernewsgithubprojects