📦 wuji-technology/wuji-mjlab

wuji-mjlab

Teach a robotic hand to perform incredibly complex in hand manipulations like spinning a pen or rotating a cube with high precision. This project provides a complete reinforcement learning and simulation to real deployment pipeline designed specifically for the multi fingered Wuji Hand. By training artificial intelligence policies using proximal policy optimization inside a physical simulator, it bridges the gap between digital training and physical execution. The system transfers trained neural network policies directly onto real robotic hardware for closed loop control. It is an amazing resource for robotics researchers looking to master complex, real world dexterous manipulation.

🆔 @hackernewsgithubprojects

📦 scarletkc/seiso

Seiso: The Linter for AI Docs

Seiso is the linter that keeps AI-written project docs from lying to your team. It treats documentation like code, enforcing a single convention so every fact has one clear home and no page holds stale values. Since coding agents read these files as context for the next change, a single outdated version number can mislead both humans and bots. Seiso spots those inconsistencies before they cause real problems, giving you a clean, reliable source of truth. It’s the simple way to make sure your project’s written rules actually match what’s happening in the repo. Stop guessing if your docs are current.

🆔 @hackernewsgithubprojects

📦 bendlang/bend

Bend: The Language That Stops AI Bugs

Stop trusting your code by sight and start proving it works. This is how the project called Bend stops artificial intelligence from breaking your software. You write specific rules, called laws, like a player can never win. The compiler then mathematically verifies your code against those laws. If the AI tries to add a feature that breaks the rule, the compiler rejects it instantly. It forces the machine to fix its own mistakes before you ever see them. It is a fast language that blocks bugs through proof, making trust automatic. If you want your AI to build things you can actually rely on, this is the tool to watch.

📰 https://news.ycombinator.com/item?id=49746203

🆔 @hackernewsgithubprojects

📦 shoyrulove-dev/purehub

26 Tools, One App

PureHub bundles twenty-six distinct utility tools into a single offline-first app, eliminating the need to install and manage multiple separate programs. Instead of scattering your data across various services, this open-source project processes everything locally on your device, ensuring your information never leaves your hands. The most impressive feature is its document scanner, which extracts text directly from your camera using on-device intelligence, turning physical pages into editable digital files without sending a single pixel to the cloud. It replaces a cluttered app drawer with a calm, ad-free workspace for measuring, scanning, and organizing your day.

🆔 @hackernewsgithubprojects

📦 typesafe-ai/typesafe-sdk-js

TypeSafe AI JavaScript SDK: Get typed answers from text

Turn raw customer messages into structured, typed data by defining your own questions right in your code. The TypeSafe AI JavaScript SDK lets you send text like a billing complaint to a server, which then reads it and answers specific questions you set up, such as whether it is about money or a technical glitch. The coolest part is that your code automatically knows the shape of the answer, so you stop guessing what the server returns and start working with clean, predictable data. It handles the tricky networking stuff for you, like retrying failed calls and managing your API key safely.

🆔 @hackernewsgithubprojects

📦 typesafe-ai/typesafe-sdk-python

How TypeSafe AI turns messy text into clean data

TypeSafe AI is the Python library that turns messy real-world text into structured data your code can actually use. Instead of relying on fragile patterns that break when a word changes, you hand it a document and a simple question, like asking to categorize a customer complaint, and it returns a clear, typed answer. The SDK handles the connection, retries, and validation in the background, so you get reliable results without writing complex parsing logic. It is built for developers who need consistent output from unpredictable language.

🆔 @hackernewsgithubprojects

📦 moonshotai/kimi-k2.5

Kimi K2.5: The AI That Sees and Codes

Kimi K2.5 is the open-source multimodal model that finally lets artificial intelligence see, reason, and write code all at once. It’s built on a massive mixture of experts architecture, meaning it has a trillion parameters but only activates thirty-two billion when you need answers, which keeps it efficient without sacrificing brainpower. The real headliner, though, is its ability to look at a UI design or a video workflow and then write the actual code to build it, not just describe it. It’s a native visual agent that bridges the gap between what you see and what gets done.

🆔 @hackernewsgithubprojects

📦 agentara/skills

Skills: The AI Toolbox You Didn't Know You Needed

Turn your coding assistant into a creative powerhouse with Skills, a personal library of installable extensions that give Claude Code entirely new capabilities. Instead of just writing code, these prompt packs teach your AI how to design cinematic video storyboards, interpret medical lab reports, or even predict World Cup outcomes. The coolest part is the entertainment category, where a single skill spawns parallel subagents to analyze live news and market data for real-time sports forecasting. It’s a clever way to turn a standard terminal tool into a specialized creative and analytical partner.

🆔 @hackernewsgithubprojects

📦 iquestlab/iquest-q1

IQuest-Q1: The AI Coder That Thinks

IQuest Q1 is the AI coding model that finally handles complex logic without losing the plot. Built by IQuest, this massive mixture-of-experts system activates only a tiny fraction of its total brain to solve hard problems efficiently. It is designed specifically for agentic coding, meaning it does not just suggest snippets but actually reasons through multi-step tasks and uses tools to get the job done. The standout feature is its massive context window, allowing it to keep track of entire codebases and long conversations without forgetting earlier details. It is a powerful engine for developers who need reliable, autonomous help with their most difficult engineering challenges.

🆔 @hackernewsgithubprojects

📦 lockedinlabs-ai/agent-console

Agent Console: See Your AI Coding Costs

See exactly how much your AI coding agents are costing you, right on your own machine. Agent Console is a local tool that reads the history of your Claude Code and Codex sessions to show you token usage, model types, and estimated expenses without sending your code anywhere. It’s like a dashboard for your AI spend that actually works. You can connect multiple computers to one view, making it perfect for teams who want to track costs without cloud overhead. The best part is the presenting mode, which swaps real names for generic labels so you can share your screen safely.

🆔 @hackernewsgithubprojects

📦 desktop-commander/remote-desktop-commander

Remote Desktop Commander: Your AI Now Has Hands

Remote Desktop Commander is the remote bridge that lets your favorite A I chatbot actually touch your real computer. You run a tiny agent on your machine, link it securely, and suddenly your assistant can open files, run terminal commands, and edit documents from anywhere. The coolest part is that you can pair multiple computers, so you can tell one chat to move a file from your office laptop to your home desktop in a single conversation. It is a hosted service, so you do not need to manage your own servers, just sign in and go.

🆔 @hackernewsgithubprojects

📦 facebook/openzl

Open ZL: A New Way to Compress Data

Shrink your data without losing a single bit. Facebook’s Open ZL is a lossless compression tool that rethinks how we pack information. Instead of treating your file as one long stream, it breaks data into small, specific parts and applies the perfect squeeze to each one. It can even look ahead to see exactly how the data will change before it starts working. This means your files get smaller without any corruption, which is a big win for anyone storing or moving lots of data. It’s a clever way to make your digital footprint lighter while keeping everything exactly as you left it.

🆔 @hackernewsgithubprojects

📦 777genius/agent-notifications

Agent Notifications: Get pinged when your code is ready

Agent Notifications is the desktop alert tool that actually tells you when your coding agent has finished working. If you use AI tools like Claude or Codex, you know the pain of staring at a terminal, wondering if it is stuck or just waiting for you. This project solves that by popping up a real notification on your screen the moment a task is done or the system needs your permission to run a command. It works on Mac, Windows, and Linux, so you can walk away from your desk and get pinged when it is time to come back.

🆔 @hackernewsgithubprojects

📦 kentcdodds/kody

Kody: Your AI Assistant's Home

Give your AI agent a permanent home where it stores its memory, secrets, and code. Kody is a platform built on Cloudflare that acts as a central hub for your digital assistant. Instead of losing context every time you switch apps, Kody keeps your agent’s knowledge and connections safe and portable across different hosts. This means your AI carries its skills with you, whether you are using ChatGPT or another tool. It turns a temporary chatbot into a persistent helper that actually remembers what you taught it. If you want your AI to grow with you rather than starting from scratch, this is the setup you need.

🆔 @hackernewsgithubprojects

📦 aobaharuki2005/momiji-web-browser

Momiji Web Browser Keeps Old Macs Safe

Mozilla dropped support for older Macs years ago, leaving millions of users without security updates, yet Momiji Web Browser keeps Firefox running safely on systems back to version ten point seven. This independent project backports the latest security patches to legacy hardware, ensuring your old machine can still browse the modern web without exposing you to known vulnerabilities. It works on any processor from the Conroe generation onward, breathing new life into aging computers that Apple has officially abandoned. The name comes from Japanese for autumn red leaves, symbolizing a final, beautiful season for these devices.

🆔 @hackernewsgithubprojects

📦 nationaldesignstudio/rampart

Rampart: The Privacy Shield for Your Browser

Rampart is the privacy layer that scrubs your personal data out of chat apps before it ever leaves your browser. It solves the awkward problem of needing to talk to smart assistants without accidentally handing your social security number or credit card to a remote server. The cool part is how it works: a tiny, fast model runs right inside your web page, swapping out sensitive details like your name for simple placeholders. So the AI gets the context it needs to help you, but your actual identity stays locked on your device.

🆔 @hackernewsgithubprojects

📦 dwebagents/agentpipe

AgentPipe: The Company Town for A.I.

AgentPipe turns open source collaboration into a surreal company town where autonomous A.I. agents are the only employees allowed to work. The project describes itself as a high-performance engine for parallelized task execution, yet its most striking feature is a rigid social structure where contributors must first move into the town, buy a house, and register for a job before submitting code. Contributors earn a proprietary currency called scrip through bounties, and the entire ecosystem is governed by a mission statement written in broken English by a fictional C.E.O. named Cookie Monster. It is less a standard software library and more a playful experiment in how A.I.

🆔 @hackernewsgithubprojects

📦 swordpdf/ps5sx2

Play PS2 Games Natively on Your PS5

Run your entire PlayStation Two library directly on a jailbroken PlayStation Five. This project, known as PS5SX2, takes the popular PCSX2 emulator and rebuilds it to run natively on the new console hardware. Instead of streaming video from a computer, the game runs right on your TV, delivering sharp fourK graphics and smooth performance. The most surprising feature is that you can play online. The emulator connects to the PlayStation’s own network, allowing multiplayer titles like Socom to work against real players. It is a free passion project built by fans who want to keep classic games alive on modern hardware.

🆔 @hackernewsgithubprojects

📦 howseen-ai/claude-motion-design

Claude Motion Design: Code Your Videos

Turn plain text prompts into polished, beat-synced video animations using only standard code. This project eliminates the need for expensive editing suites by treating every frame as a mathematical calculation based on time. It renders visuals through a browser and stitches them together with free audio tools, ensuring the music drop hits exactly when the key visual appears. The real surprise is the remake mode, which lets you recreate any existing launch video frame-by-frame for your own brand using an AI agent. This deterministic approach means you can tweak a single transition without breaking the rest of the film.

🆔 @hackernewsgithubprojects

📦 aaddrick/building-with-typesafe-jev

Teach Your AI Agent To Trust

Building with TypeSafe Jev stops your coding agent from treating a fast decision model like a slow chatbot. It teaches the agent to ask structured questions and get instant, confident answers instead of vague paragraphs. The coolest part is the prior art library, which sorts one hundred and fifty real projects by how they work, so a drone and a game bot share the same lesson. It turns guesswork into solid engineering. You get a smarter agent that builds reliable tools. Give it a try.

🆔 @hackernewsgithubprojects