📦 kmx415/meshpoint

Meshpoint

Meshpoint is the open-source base station software that finally turns cheap, surplus Helium miners into ultimate Meshtastic powerhouses. By pairing a Raspberry Pi with a multi-channel concentrator, this clever setup lets you monitor and transmit across eight channels simultaneously instead of relying on a tiny, single-channel pocket radio. You get a gorgeous local web dashboard to chat, manage direct messages, and track every node in range on a live map. It even acts as an identity-preserving smart relay to help your local network grow. Just flash your Pi, load the dashboard, and dominate the airwaves.

🆔 @hackernewsgithubprojects

📦 espressif/esp-claw

ESP-Claw

Espressif has built an artificial intelligence agent framework that runs entirely on a tiny microchip costing just a few dollars. The project is called esp-claw, and it brings active decision-making directly to edge hardware like the ESP32. Instead of relying on rigid, pre-programmed code, this framework allows you to define and change your hardware's behavior simply by chatting with it through messaging apps like Telegram. It translates your text instructions into Lua scripts, downloads them, and runs them locally on the device in milliseconds. It turns simple silicon into a smart, self-programming assistant that keeps your private data off the cloud.

🆔 @hackernewsgithubprojects

📦 facebookresearch/autopartgen

AutoPartGen

You can now automatically break down a single flat image or a solid 3D mesh into a complete set of individual, fully functional 3D part meshes. The open source project autopartgen makes this possible by using an autoregressive approach to predict and discover the distinct components of an object. Instead of treating a 3D model as one solid block, it segmentally generates watertight individual parts like the petals of a flower or the limbs of a robot. It is an incredibly clever way to turn static shapes into ready to animate assets. Check out the repository to start decomposing your own 3D designs today.

🆔 @hackernewsgithubprojects

📦 evolvinglmms-lab/lmms-lab-writer

lmms-lab-writer

Write academic papers with AI without ever leaking your drafts or data. The lmms-lab-writer repository provides a local-first desktop LaTeX editor that puts privacy and productivity first. It automatically configures a lightweight LaTeX setup on your machine, eliminating painful background installations. You can chat directly with your choice of artificial intelligence models to write sections, edit text, and organize references in real-time. It also includes full offline capabilities and built-in Git tools for painless version tracking. It is a brilliant way to write beautiful research papers with smart assistance while keeping complete control over your intellectual property.

🆔 @hackernewsgithubprojects

📦 andymik90/aperant

Aperant

Hand over your entire coding to-do list to a team of digital developers and watch them build your software while you drink coffee. This desktop app called aperant connects directly to your git repository to coordinate multiple autonomous agents that work together. Instead of copy-pasting code snippets back and forth, you just describe the task, and the system automatically plans the architecture, writes the code, and runs tests to validate everything. It turns your development process into a visual board where you can watch tasks move from planning to completed. Grab the desktop app and let your new digital team handle the heavy lifting.

🆔 @hackernewsgithubprojects

📦 shootthesound/fizgig

Fizgig

Fix broken image generation adapters block-by-block without ever restarting a training run. This clever workbench is designed specifically for fine-tuning newer massive AI models on normal home computers. Instead of throwing away a model that overbakes details or crushes your style, you get thirty-two interactive sliders to surgically dial back the influence of individual transformer layers. It even lets you evolve model variations like a mini-game by picking your favorite mutations. You can easily find the perfect checkpoint, mix different styles, and export smooth visual timelapses to share. It is the ultimate playground for customizing your generative artwork.

🆔 @hackernewsgithubprojects

📦 blue-pen5805/comfyui-krea2-negpip

ComfyUI Krea 2 NegPiP

ComfyUI Krea 2 NegPiP is the custom node that finally brings precise negative prompt weighting to your Krea 2 image generation workflows. When creating AI art, telling the generator what to avoid can be incredibly tricky. This clever tool solves that by allowing you to use negative numbers directly inside your standard prompt text, letting you easily suppress unwanted elements like blurry details or low quality. By patching into your model and clip setup, it gives you a simple slider to control how strongly those negative cues are filtered out. It is a brilliant, straightforward way to clean up your generations with incredible precision.

🆔 @hackernewsgithubprojects

📦 faizkhairi/claude-code-blueprint

Claude Code Blueprint

Prevent your AI from making costly, repetitive programming mistakes by grabbing a battle-tested reference architecture designed specifically for Claude Code. This clever setup gives you a library of ready-to-copy configuration files, behavioral rules, and automation hooks that you simply drop into your existing workspace. The absolute best part is how it uses path-scoped rules to keep your token costs low, only loading specific instructions when you actually edit matching files. It is the ultimate way to keep your sessions incredibly efficient while making your AI assistant significantly more reliable. Copy a single file to get started and see the difference in your very next session.

🆔 @hackernewsgithubprojects

📦 langchain-ai/openwiki

OpenWiki

Your codebase is changing constantly, but your documentation is probably stuck in the past. OpenWiki is a clever command-line tool that uses intelligent agents to automatically write, structure, and maintain deep documentation for your entire repository. Instead of guessing how your app works, it scans your git history, inspects your files, and creates a perfectly organized wiki right in your project folder. It even hooks into your git workflows to open daily pull requests with fresh updates only when your code actually changes. It is like having a dedicated technical writer who never sleeps, keeping your team and your coding agents perfectly aligned.

📰 https://news.ycombinator.com/item?id=48752949

🆔 @hackernewsgithubprojects

📦 rh-aiservices-bu/s4

s4

Spin up a lightweight, fully functional object storage server locally with just a single container run. When you are building prototypes, demos, or local development environments, setting up a massive storage setup is total overkill. This project, called s4, solves that by combining a simple storage API with a built-in web browser for your files, all backed by a standard filesystem instead of a massive database. The coolest part is that you can manage your buckets, trigger webhook notifications when files change, and even import machine learning models directly from HuggingFace. It gives you a complete sandbox to test storage code on your laptop before deploying.

📰 https://news.ycombinator.com/item?id=48081582

🆔 @hackernewsgithubprojects

📦 alekk89/llama-cpp-windows-manager

llama-cpp-windows-manager

The llama-cpp-windows-manager is the desktop console that finally makes running local AI models on Windows painless. Instead of wrestling with complex command-line arguments and manual environment setups, this slick desktop application coordinates everything for you right on your PC. It lets you download models directly from Hugging Face, manage different runtimes, and spin them up with customized launch profiles in a couple of clicks. The coolest part is how it integrates with both native Windows and WSL, giving you a smooth way to build, run, and monitor your local model APIs for coding workflows without any of the usual headache. Give it a spin and see how easy local AI...

🆔 @hackernewsgithubprojects

📦 christopherarter/superpowers-reasonix

Superpowers Reasonix

Superpowers-reasonix is the coding-agent plugin that finally brings rigorous, human-grade engineering discipline to your AI workspace. Instead of letting your AI model rush ahead to write messy code, this clever harness forces it to stop, plan, and think first. It ports a set of eleven essential software disciplines to the Reasonix platform, teaching the AI to write tests before implementing features, trace root causes before debugging, and run thorough verifications before claiming a task is done. It even includes a complete benchmarking suite to verify the model actually obeys these instructions. It is the ultimate way to turn a fast but sloppy AI assistant into a careful, reliable peer.

🆔 @hackernewsgithubprojects

📦 heymrun/heym

Heym AI Native Workflows

Build and run complex AI agents on a visual canvas without writing a single line of code. Heym changes how we automate tasks by putting language models at the very center of the execution rather than just plugging them in as an afterthought. You can visually wire together smart agents, web scrapers, and data tables to build workflows that actually talk to each other. The coolest part is that you can build these entire automated systems using just your voice or plain text prompts, and the visual editor automatically handles the wiring. It runs entirely on your own hardware, so your data stays private and secure.

🆔 @hackernewsgithubprojects

📦 hkuds/paper2slides

Paper2Slides

Turn dry research papers, PDFs, and Word documents into fully designed, professional presentation slides and academic posters with a single command. It uses an intelligent four-stage pipeline that parses your files, extracts key figures, plans a structured layout, and automatically generates high-quality visuals. The coolest part is how easily you can customize the aesthetic. Instead of wrestling with formatting, you can describe your design ideas in plain English, and the system renders customized slides to match your theme perfectly. It is a fantastic open-source tool for turning complex research into beautiful, ready-to-share presentations instantly.

🆔 @hackernewsgithubprojects

📦 showlab/paper2video

Paper2Video

You can now transform a dense LaTeX academic paper into a fully animated presentation video complete with voiceover, slides, active cursor tracking, and a speaking virtual presenter. This project makes complex research accessible by automating the entire pipeline. It takes your raw scientific paper, a portrait photo, and a quick audio clip to generate a polished lecture where a digital avatar explains your findings while pointing to key details on the screen. It takes the pain out of research promotion, giving scientists an easy way to share their hard work with the world. You simply feed it your draft, and you walk away with a ready-to-share video.

🆔 @hackernewsgithubprojects

📦 yashab-cyber/hackgpt

hackgpt

You can now run a full-scale enterprise security audit using a voice command on your phone. This is possible because hackgpt connects advanced artificial intelligence with a massive suite of network scanning tools to find and fix security weaknesses in large systems. Instead of forcing security teams to manually run dozens of separate programs, it orchestrates the entire process and automatically drafts professional compliance reports. The coolest part is that it can even take spoken instructions to spin up workers or trigger a test on the fly. It is like having a seasoned security analyst on standby, ready to protect your servers whenever you ask.

🆔 @hackernewsgithubprojects

📦 agentskills/agentskills

agentskills

agentskills is the open standard that finally gives AI assistants a clean, portable way to learn new jobs. Originally created by Anthropic, it lets you pack custom instructions, scripts, and templates into a simple folder with a markdown file. Instead of cluttering your AI's memory with a wall of instructions, it uses a clever system where the assistant only peeks at the skill's name and description at first. It only loads the full, heavy workflow instructions when it actually needs to execute that specific task. This keeps things incredibly fast and lightweight. It is the perfect way to build reusable, context-aware AI tools that actually work.

🆔 @hackernewsgithubprojects

📦 jackwener/wx-cli

wx-cli

Your local WeChat message history is actually fully readable on your computer without sending a single byte of data to the cloud. A brilliant Rust-based command-line tool called wx-cli bypasses typical database headaches by scanning your running WeChat process memory to extract the active SQLCipher key. From there, a background helper process securely decrypts and caches your databases locally. This lets you instantaneously search through group chats, look up contacts, extract encrypted images, or even feed your raw chat history directly into local AI models and agents. It keeps your data completely offline and lightning-fast.

🆔 @hackernewsgithubprojects

📦 microsoft/skills

skills

The skills project is the developer toolbox that feeds structured knowledge directly to coding assistants to make them write better code. While standard AI assistants struggle with outdated library versions and API changes, this system organizes code blueprints, SDK examples, and architectural patterns into bite-sized units of knowledge. It uses the Model Context Protocol to instantly ground AI agents in real, up-to-date SDK details across languages like Python, Java, and Rust. Instead of guessing how to implement a complex cloud integration, your coding assistant reads these curated blueprints to construct correct solutions. It is a smart way to keep your autonomous agents accurate and production-ready.

🆔 @hackernewsgithubprojects

📦 modelcontextprotocol/ext-apps

ext-apps

ext-apps is the model context protocol extension that finally brings interactive user interfaces to your favorite AI chatbots. While traditional AI assistant tools are stuck outputting basic text and raw data, this clever SDK lets your servers serve up full, interactive web views like charts, design canvases, and live dashboards directly inside the chat window. It works by standardizing how a chat client renders sandboxed web apps in real time, letting the bot and the UI talk back and forth seamlessly. You can even have an AI coding agent build these interfaces for you from scratch using the included agent skills. It is the perfect way to turn static text...

🆔 @hackernewsgithubprojects