📦 milind-soni/openmausbot

OpenMausBot: Your AI Team in a Chat App

Manage your entire AI workforce from a simple chat window. OpenMausBot transforms your existing AI tools into a roster of interactive contacts, letting you message them like people. The standout feature is that these bots can actually use a computer, browsing files and running tasks while you watch their screen live. It’s an open-source, local-first alternative to private AI assistants, giving you full control over who works, which model they use, and exactly what permissions they have. You stay in the loop, approving every risky action before it happens. It’s a practical, hands-on way to supervise your digital team without losing control.

🆔 @hackernewsgithubprojects

📦 cpj-dev/dsh-plugin-cc

dsh-plugin-cc: The Second Opinion Plugin

dsh-plugin-cc is the bridge that lets you ask DeepSeek to critique your work directly inside Claude Code. Usually, when you are coding with an AI assistant, you are trusting a single model to handle everything. But what if you wanted a completely different brain to double-check your logic? This plugin acts as a translator, allowing you to run powerful, independent review tasks using DeepSeek without ever leaving your current workflow. Think of it as hiring a second expert to look over your shoulder while you work. You can hand off complex tasks, get structured adversarial feedback, or review your local changes instantly.

🆔 @hackernewsgithubprojects

📦 thebuggeddev/seed

Seed: 3D Crop Science Atlas

Seed is an interactive 3D atlas that turns the inside of eight staple crops into a hands-on science lab. The most surprising part? The app doesn't just show you a seed; it slices it open, germinates it on screen, and lets you click specific tissues to see exactly how the plant works. Built with three.js, it uses clever lighting and custom shading to make things like a glossy bean or matte wheat look physically real, not just like a 3D model. You can rotate, zoom, and even take quizzes to test your anatomy knowledge. It’s a brilliant way to turn a tiny piece of nature into a deep-dive learning experience.

🆔 @hackernewsgithubprojects

📦 thebuggeddev/seed

Seed: 3D Crop Science Atlas

Seed is an interactive 3D atlas that turns the inside of eight staple crops into a hands-on science lab. The most surprising part? The app doesn't just show you a seed; it slices it open, germinates it on screen, and lets you click specific tissues to see exactly how the plant works. Built with three.js, it uses clever lighting and custom shading to make things like a glossy bean or matte wheat look physically real, not just like a 3D model. You can rotate, zoom, and even take quizzes to test your anatomy knowledge. It’s a brilliant way to turn a tiny piece of nature into a deep-dive learning experience.

🆔 @hackernewsgithubprojects

📦 thebuggeddev/seed

Seed: 3D Crop Science Atlas

Seed is an interactive 3D atlas that turns the inside of eight staple crops into a hands-on science lab. The most surprising part? The app doesn't just show you a seed; it slices it open, germinates it on screen, and lets you click specific tissues to see exactly how the plant works. Built with three.js, it uses clever lighting and custom shading to make things like a glossy bean or matte wheat look physically real, not just like a 3D model. You can rotate, zoom, and even take quizzes to test your anatomy knowledge. It’s a brilliant way to turn a tiny piece of nature into a deep-dive learning experience.

🆔 @hackernewsgithubprojects

📦 thebuggeddev/seed

Seed: 3D Crop Science Atlas

Seed is an interactive 3D atlas that turns the inside of eight staple crops into a hands-on science lab. The most surprising part? The app doesn't just show you a seed; it slices it open, germinates it on screen, and lets you click specific tissues to see exactly how the plant works. Built with three.js, it uses clever lighting and custom shading to make things like a glossy bean or matte wheat look physically real, not just like a 3D model. You can rotate, zoom, and even take quizzes to test your anatomy knowledge. It’s a brilliant way to turn a tiny piece of nature into a deep-dive learning experience.

🆔 @hackernewsgithubprojects

📦 priorlabs/relarena

RelArena: The Ultimate Test for Relational AI

Stop guessing which AI model actually handles messy, connected data. RelArena gives you a single, clean stage to pit every major relational learning method against each other on a shared set of real-world problems. The biggest headache in this field is that everyone tests their model differently, making results impossible to compare. This project fixes that by locking in the same data, the same tuning rules, and the same evaluation standards for everything. You can throw graph neural networks, table-based models, or brand-new ideas into this ring and see who truly wins.

📰 https://news.ycombinator.com/item?id=49363970

🆔 @hackernewsgithubprojects

📦 erroralex/latent-library

Organize Your AI Image Chaos with Latent Library

Sort through thousands of AI-generated images without losing your mind. Latent Library is a desktop tool that turns your chaotic folder of art into a searchable, organized library. It reads the hidden details saved in your image files, telling you exactly which model or settings created each piece, so you can find that perfect shot instantly. The coolest part is its local AI tagging, which automatically labels your images using a built-in model that runs entirely on your computer, keeping your data private. It even spots duplicate images so you can clean up your hard drive with ease.

🆔 @hackernewsgithubprojects

📦 pythops/kudu

Kudu: A Terminal VM Manager for Linux

Kudu runs entirely in your terminal, turning complex virtual machine management into simple keyboard navigation. Most Linux virtual machine managers require a graphical desktop and heavy dependencies, but Kudu bypasses this entirely. It is a text-based interface that talks directly to QEMU, meaning it runs even on bare-metal servers without a display. The tool handles the messy details of networking, storage, and firmware setup automatically, letting you spin up new virtual machines or edit existing ones with just a few key presses. It also integrates cloud-init to generate the necessary configuration disks for fresh installs, saving you from manual command-line tweaking.

📰 https://news.ycombinator.com/item?id=49425330

🆔 @hackernewsgithubprojects

📦 fzakaria/selfdb

Your Program Is Just A Database Now

Selfdb is the executable format that turns your compiled program into a standard SQLite database file. Normally, a program is a complex binary blob that requires specialized tools to inspect or modify, but here, the code itself lives inside simple database rows. This means you can use regular SQL queries to see what libraries a program needs or even strip out debug information with a basic delete command, and the program still runs perfectly. By treating software as structured data, this project makes binary analysis feel as easy as querying a spreadsheet, proving that the most complicated parts of your computer can just be rows in a table.

📰 https://news.ycombinator.com/item?id=49416878

🆔 @hackernewsgithubprojects

📦 tonbistudio/hermes-desktop-plugins

Play SNES Inside Your AI App? Hermes Desktop Plugins

Hermes Desktop Plugins is the open-source toolkit that lets you drop custom tools directly into your AI workspace without touching the main app. It’s like adding new tabs to your browser, but for an AI agent. The coolest part is the SNES emulator plugin. You just point it at a local ROM file on your computer, and suddenly you have a fully working Super Nintendo console right inside a sidebar pane of your AI assistant. Your game session stays alive in the background even if you switch to writing code or checking emails. It’s not just about gaming, though.

🆔 @hackernewsgithubprojects

📦 shdwmtr/wallpiper

Wallpiper: Linux Wallpapers, Finally Native

Wallpiper is the lightweight tool that finally brings real Wallpaper Engine animations to Linux without the bloat. Instead of cloning the whole app, it acts as a tiny bridge that lets the original Windows version run smoothly under Proton. It then pipes the visual output directly into your desktop, supporting major systems like Hyprland, Sway, and GNOME. The coolest part is how efficient it is. The entire core weighs in at just one hundred kilobytes, proving that you don't need a massive re-implementation to get a working solution. You get native performance and full multi-monitor support.

📰 https://news.ycombinator.com/item?id=49412578

🆔 @hackernewsgithubprojects

📦 water-run/luainstaller

luainstaller: Turn Lua Scripts Into Binaries

One surprising fact is that luainstaller packages a standard Lua script into a single binary executable that runs without any system Lua interpreter installed. It solves the distribution headache by bundling your code, dependencies, and even native modules into a portable package, much like how Python developers use PyInstaller. The tool automatically discovers required files and generates a self-contained artifact, so users don't need complex environment setups to run your application. This makes sharing Lua projects significantly easier and more reliable.

🆔 @hackernewsgithubprojects

📦 colorado-mesh/mesh-client

One App for Three Mesh Networks

One single desktop app now lets you chat and see maps across three totally different mesh radio systems at once. Mesh Client brings Meshtastic, MeshCore, and Reticulum into one window, so you can switch protocols without juggling separate tools. It handles the messy bits of hardware connection, like Bluetooth or USB, and keeps your message history safe locally. The real surprise is the unified view: you can hop between networks instantly while tracking where your messages actually went. This makes offline communication way less intimidating for anyone setting up a resilient network. Next time your phone dies, you have a plan that works.

📰 https://news.ycombinator.com/item?id=49174086

🆔 @hackernewsgithubprojects

📦 schlarpc/re-shell

Re-Shell: The AI-Powered Reverse Engineering Kit

Re-Shell is the reverse engineering environment that finally lets AI handle the tool setup for you. Instead of spending hours configuring static analyzers and decompilers, you just drop a binary into the folder and ask Claude Code to investigate. The system automatically detects what kind of file you have, like a Windows executable or an Android package, and instantly activates the right set of tools for the job. It even bundles in heavy hitters like Ghidra and Frida so you have everything ready to go. The coolest part is that it can build itself.

📰 https://news.ycombinator.com/item?id=49413569

🆔 @hackernewsgithubprojects

📦 laude-institute/headlong

Headlong: The AI That Thinks in Bash

Headlong is the agent microharness that finally lets an AI think on its own. Most AI tools just wait for you to type a command, but this one runs a constant loop, keeping a stream of thoughts even when you are not talking to it. You give it a name, and it starts projects on its own. The coolest part is that it doesn't use fancy code libraries to do work. It simply writes shell commands, runs them, and reads the results, treating the terminal like a human operator would.

🆔 @hackernewsgithubprojects

📦 tobi/walgit

One Binary, Infinite Repos

Run a full git server on hardware that cannot possibly hold your code, and lose nothing if the machine dies. Walgit is a single binary that points to a cloud storage bucket, treating that bucket as the only source of truth. Every machine running it is just a disposable cache, so you can scale to repositories larger than your server without complex databases or leader elections. The coolest part is how clones work: instead of the server streaming gigabytes of data, it hands out static bundles directly from the bucket, letting a content delivery network do the heavy lifting.

📰 https://news.ycombinator.com/item?id=49420598

🆔 @hackernewsgithubprojects

📦 tjqscott/dory

Dory: The AI Assistant That Remembers Without the Bloat

Dory treats your hosted AI chats as disposable tools by moving all your project memory to your local hard drive instead of letting it live in the browser. The biggest problem with long coding sessions is that every new message forces the AI to re-read the entire history, which burns through your token limits and makes the model forget earlier details. Dory solves this by saving your code decisions and session summaries into a local folder within your project. This means you can close your browser tab at any time without losing progress, and when you start a new chat, you simply load that saved history.

🆔 @hackernewsgithubprojects

📦 adobe-research/ldr

LDR: Predicting the Future of Video

Take just three frames of a video and LDR predicts how the entire scene evolves into the future. This Adobe and UCSD project acts like a video world model that doesn't just mimic pixels, but actually understands physical rules. It figures out how objects move, collide, or bounce based on their hidden momentum. Instead of simply guessing the next picture, it uses latent dynamics reasoning to extrapolate realistic outcomes far beyond what you see. It works across tasks like uniform motion or collisions. This is a huge step for AI that truly comprehends how our world physically evolves over time.

🆔 @hackernewsgithubprojects

📦 nguyenphutrong/quotio

Quotio: The AI Account Manager for Mac

A tiny app in your Mac menu bar quietly switches between your AI coding accounts the moment one runs out of tokens. Quotio acts as a smart router for tools like Claude Code and OpenCode, keeping your work moving without constant interruptions. It connects multiple AI subscriptions, monitors their limits in real time, and automatically fails over to a fresh account when the current one hits its limit. This stops the annoying mid-code error messages that force you to manually swap providers. It’s the missing middleman that keeps your coding agents running smoothly. Now your AI tools just work, no matter which account is available.

🆔 @hackernewsgithubprojects