📦 gimi-xiaomi/gimi-illustration-skill

gimi-illustration-skill

gimi-illustration-skill is the open-source AI agent skill that transforms your written articles into beautiful, quirky hand-drawn illustrations. If you write blog posts, newsletters, or presentations, this tool automatically reads your Chinese text, analyzes the core ideas, and generates high-quality visual metaphors in a cozy sketch style. It supports a charming default character, a clean no-character mode, or you can even upload and lock in your own custom brand mascot. By handling everything from composition strategy to quality checks, it gives you perfectly framed, story-driven visuals that make your written ideas instantly shareable and easy to understand.

🆔 @hackernewsgithubprojects

📦 capitalone/vulnhunter

VulnHunter

VulnHunter is an open-source security tool that thinks like an actual attacker to find real security flaws in your source code. Instead of acting like traditional scanners that flood your screen with fake warnings, it flips the script. It starts at public-facing entry points, like APIs or file uploads, and traces forward to see if a hacker could actually break in. It even uses a clever falsification engine to double-check itself, trying to disprove its own findings before bothering you. When it finds a real bug, it writes a proof-of-concept exploit, patches the code, and verifies the fix is safe.

📰 https://news.ycombinator.com/item?id=48942583

🆔 @hackernewsgithubprojects

📦 gintasz/unigent

Unigent

Unigent is the cross-harness SDK that finally lets you build stateful agent workflows using ordinary TypeScript. Instead of wrestling with complex frameworks or manual string parsing, you write simple functions and let the SDK automatically extract your TypeScript signatures to generate clean tool definitions. You can spin up conversational agents, effortlessly fork sessions to explore different paths simultaneously without duplicating context, and get strictly typed outputs directly from your models. There is even a live trace terminal inspector to watch your agents work in real-time. Give it a spin to build smarter, self-correcting automation scripts today.

📰 https://news.ycombinator.com/item?id=48947117

🆔 @hackernewsgithubprojects

📦 scrapfly/antibot-detector

antibot-detector

antibot-detector is the Chrome extension that reveals exactly how websites spy on your browser and block automated visitors. Instead of guessing why a page is blocking you or forcing a CAPTCHA, this tool works right inside your browser to intercept and expose security scripts in real time. It monitors everything from network requests and cookies to hidden JavaScript hooks that silently collect your canvas, audio, and hardware fingerprints. You get a clear confidence score showing which security networks are active on any page you visit. It is a brilliant way to understand modern web security and see how the sites you visit analyze your digital identity.

🆔 @hackernewsgithubprojects

📦 fujacob/cotabby

Cotabby

Your Mac can now autocomplete your writing in almost any app without ever sending a single word to the cloud. A clever open-source tool called cotabby brings system-wide, inline ghost-text suggestions directly to your screen, working in everything from Slack to iMessage. It uses your choice of Apple's built-in models or local open-source models running entirely on your device. That means you get lightning-fast predictions, emoji completions, and smart autocorrect with absolute privacy and zero tracking. It is a brilliant way to speed up your daily typing, keeping your data safely on your machine where it belongs.

🆔 @hackernewsgithubprojects

📦 data-centt/percentify

percentify

Most data-profiling tools bury you in massive, hard-to-read reports, but percentify takes a completely different approach by giving you a single-line data diagnostician that tells you exactly what to fix before you model. Instead of just describing what your dataset is, its flagship profiler function ranks every data issue from worst to first and pairs each with a concrete suggestion on how to solve it. It seamlessly handles both pandas and Polars dataframes, automatically returning whichever type you passed in without requiring any annoying manual conversions or configuration flags. It is like having a sharp, slightly opinionated data scientist review your dataset over coffee, instantly calling out multicollinearity, outliers, and...

📰 https://news.ycombinator.com/item?id=48948400

🆔 @hackernewsgithubprojects

📦 sompote/tigrimosr

TigrimOSR

Build and run your own supercharged native desktop AI assistant with zero lag or bloat. TigrimOSR is a high-performance native desktop application written entirely in Rust that replaces the original heavy Python and Node.js setup with a single, self-contained binary. It uses egui for a lightning-fast, ultra-responsive user interface that loads instantly and keeps computer memory usage incredibly low. The coolest thing is its modular skill system, allowing the local agent to dynamically handle tasks like analyzing complex Excel student grades, auto-generating charts, and conducting deep market research using specialized workflows. It is the ultimate way to run a nimble, private AI assistant on your desktop.

🆔 @hackernewsgithubprojects

📦 thesysdev/appless

Appless

Appless is the experimental mobile shell that replaces a screen full of icons with a phone that has no apps. Instead of installing menus and software, you simply type what you want in plain words and the device generates the exact interface you need on the spot. Ask for a weather dashboard, a restaurant finder, or a workout tracker, and a real native user interface streams onto your screen in real time. It uses a lightweight interface language to paint authentic Cupertino components on iOS and Material components on Android. You get custom, on-demand screens tailored to the moment.

📰 https://news.ycombinator.com/item?id=48936362

🆔 @hackernewsgithubprojects

📦 tura-ai/tura

Tura

Build software more efficiently by skipping the endless back-and-forth loops of traditional AI assistants. Instead of sending minor commands back and forth for every single step, this local coding agent packages your entire workflow into a single turn using structured macro commands. It groups searching, patching, building, and testing into one logical execution tree, saving a massive amount of conversational overhead. It even guides the model to reason backward from the final goal to map out the exact root cause of a bug before writing a single line of code. If you want a smarter, more deliberate development assistant, check out tura.

📰 https://news.ycombinator.com/item?id=48922039

🆔 @hackernewsgithubprojects

📦 patonw/leaves

Leaves

Visualize your disk space in a clean, interactive text-mode map that runs directly inside your terminal. Taking inspiration from classic disk tools, this application structures files and directories into nested rectangles where the size of each block corresponds to its actual footprint on your drive. Working inside a remote shell with no graphical interface, you can effortlessly browse your folders, navigate with arrow keys, and zoom in on heavy directories. It dynamically groups deep directories and color-codes everything by file extension, helping you instantly find bloated folders. Give it a run to reclaim control of your storage from anywhere.

📰 https://news.ycombinator.com/item?id=48936389

🆔 @hackernewsgithubprojects

📦 dannymac180/skills

Skills

Supercharge your AI agent by teaching it how to actually help you learn and manage complex tasks. This repository contains custom skills you can install directly into your AI assistant. The standout feature is a tool called Explain This, which completely changes how you read papers, articles, or code. Instead of giving you a generic, lazy summary, it interviews you to build a custom learning profile, explains complex concepts using analogies you already understand, and then quizzes you to make sure you actually get it. It even saves flashcards for spaced-repetition review so you never forget what you learned.

🆔 @hackernewsgithubprojects

📦 google-gemma/gemma-skills

Gemma Skills

Developers can now run high-efficiency quantization-aware training models that simulate compression during training rather than after, preserving crucial intelligence on edge devices. This capability is powered by gemma-skills, a repository designed to help you build and fine-tune Gemma models locally. It acts as an interactive blueprint, helping you deploy workflows like multi-token prediction to speed up inference or use memory-efficient tools for local fine-tuning. Whether you are running lightweight models on a laptop or deploying multimodal setups with text, images, and audio, this project provides the exact recipes and templates you need to build intelligent, responsive local AI assistants.

🆔 @hackernewsgithubprojects

📦 evermind-ai/raven

Raven

Build and run AI assistants that actually remember your preferences and learn from their mistakes. While typical AI agents forget your conversations the moment you close the window, raven uses a dedicated memory backbone to keep your workflows, tools, and personal context alive across different sessions. It runs directly inside your command line terminal with a beautiful interactive interface, and it can even detect your repetitive tasks to turn them into reusable skills. This is a game-changing way to build digital helpers that adapt to your personal work environment instead of resetting to zero every single time.

🆔 @hackernewsgithubprojects

📦 kaifcodec/ytconverter

ytconverter

ytconverter is the terminal-based downloader that finally makes saving YouTube media dead simple. Instead of forcing you to remember complicated command-line arguments, this tool builds a colorful, interactive menu right in your terminal. You can download entire playlists, extract high-quality audio files, grab subtitles, and choose your exact video resolution with simple keystrokes. It automatically handles the tricky stuff behind the scenes, like cleaning up weird file names and auto-detecting your downloads folder whether you are on Windows, Linux, or mobile on Termux. It is a brilliant, menu-driven way to grab your favorite videos offline without the browser hassle.

🆔 @hackernewsgithubprojects

📦 samyeyo/clx

Compile Lua to Native Binaries with CLX

Compile your Lua scripts directly into standalone, native executables that run without any interpreter overhead. Using clx, you can take standard Lua code and compile it ahead-of-time through modern C++ compilers like GCC, Clang, or MSVC. This approach gives you lightning-fast startup times, predictable runtime performance, and incredibly compact binaries under one hundred kilobytes. By translation to optimized C++, it leverages decades of compiler development for massive speedups on intensive workloads. It is perfect for developers who love the simplicity of Lua but need the speed, portability, and easy distribution of native machine code.

📰 https://news.ycombinator.com/item?id=48871547

🆔 @hackernewsgithubprojects

📦 erickong/penecho

penecho

You can now solve complex math and sketch diagrams directly on an open digital canvas, and watch an AI read your handwriting to solve the problem right next to your pen strokes. This tool, called penecho, bridges the gap between raw brainstorming and artificial intelligence by turning your visual scribbles into structured, editable drafts. Instead of typing formulas into a boring chat box, you simply write or draw naturally with your mouse or stylus on a massive workspace. The app sends your visual layout to a model to generate a movable draft that you can resize, tweak, or accept. Grab a stylus, run a quick local command, and start sketching...

🆔 @hackernewsgithubprojects

📦 pwilkin/trellis.cpp

Trellis dot CPP

You can now transform a flat two-dimensional image into a fully textured, three-dimensional digital model with zero Python dependencies. A clever new project called trellis.cpp delivers a complete rewrite of Microsoft's advanced image-to-three-D pipeline in pure C++. By utilizing a custom neural engine and running directly on your GPU, it handles everything from automatic background removal and shape decoding to generating a high-quality three-D mesh with beautifully baked textures. It is incredibly lightweight and easily fits onto consumer graphics cards. If you want to integrate high-fidelity three-D generation directly into your native applications, this is the ultimate starting point.

🆔 @hackernewsgithubprojects

📦 bouzidanas/pynote-notebook-editor

PyNote Notebook Editor

You can now run a complete Python data science notebook directly in your web browser with zero server setup or cloud costs. This is pynote-notebook-editor, a serverless, local-first environment powered entirely by WebAssembly. Because your code runs in a sandboxed browser thread, your data never leaves your machine, making it completely private and secure. What makes it incredibly cool is its built-in interactive UI system; you can design custom web forms, reactive sliders, and polished charts directly from your Python code, and they will update in real time without you ever needing to write frontend JavaScript. It is the perfect way to build and share gorgeous, interactive reports that look...

🆔 @hackernewsgithubprojects

📦 huytieu/cog-second-brain

cog-second-brain

Your markdown files can now think for themselves. COG-second-brain turns a local folder of text files into a self-evolving knowledge engine using local AI agents and git. Instead of wasting hours organizing notes, you just dump your raw thoughts into a text file. The system automatically classifies the domain, links related topics, and extracts action items. It even tracks your professional network with a progressive people database that gets smarter the more you mention them. It works directly with tools like Claude Code and Cursor to build structured, self-healing frameworks from your scattered thoughts. It is the ultimate zero-maintenance system that actually grows with you.

🆔 @hackernewsgithubprojects

📦 vlm-run/mm

Fast Multimodal Context with mm

mm is the command-line tool that finally gives classic UNIX commands like cat, grep, and find multimodal powers. Instead of hitting a wall when your scripts encounter images, videos, audio, or PDFs, this clever tool lets you inspect and search binary formats as if they were plain text. Backed by a fast Rust core, you can instantly extract audio transcripts, scan PDF tables, or generate detailed image descriptions right from your terminal. It even packs a python API to build multimodal prompts for your favorite AI models in just a few lines. Toss it into your terminal, and let your command line finally see the world.

🆔 @hackernewsgithubprojects