📦 tt-a1i/simplify-codebase

Simplify Codebase: The AI Agent That Proves Before It Deletes

This AI agent refuses to delete a single line of code until it has mathematically proven that the system will survive the cut. Named simplify codebase, it acts as a cautious auditor for your software, hunting down the invisible clutter that quietly slows teams down, like duplicate settings or abandoned features hiding in plain sight. Instead of blindly trimming bulk, it traces exactly who uses every part of your app before making a move, ensuring you keep the critical pieces and toss only the true dead weight. It is a powerful way to shrink your mental load without breaking anything.

🆔 @hackernewsgithubprojects

📦 andreasmolnardev/dashwise

Dashwise: Your Homelab Dashboard

Dashwise is the homelab dashboard that puts your entire self-hosted world into a single, beautiful view. It solves the clutter problem by letting you drag and drop customizable widgets, links, and news feeds onto one clean page, so you stop jumping between tabs for every little update. The standout feature is its built-in search bar; just press Control and K, type a command, and instantly find any app or run a specific search engine query without leaving the screen. It handles login securely and lets you configure everything right from the interface, so you never have to mess with code files again.

📰 https://news.ycombinator.com/item?id=49522428

🆔 @hackernewsgithubprojects

📦 yu-xiao-sheng/codex-memory-trim

Codex Memory Trim: The Brain Diet

Your AI assistant is secretly getting slower because it never forgets. Codex Memory Trim solves this by performing a periodic diet on your AI's global memory. The creator found that Codex was applying outdated rules from months ago to brand new tasks, causing it to waste hours on redundant checks. This tool fixes that by finding duplicate notes, deleting dead ends, and compressing long-winded explanations into concise lessons. It acts like a librarian for your digital brain, ensuring the AI stays fast and relevant. The most surprising part is that it cuts memory size by over seventy percent while actually making the AI more efficient.

📰 https://news.ycombinator.com/item?id=49361485

🆔 @hackernewsgithubprojects

📦 kaomei/stickman-video-director

Stickman Video Director: The AI Director for Your One-Minute Videos

Stickman Video Director is the creative workflow that turns plain text into a fully directed sixty-second stickman animation. Most AI video tools give you random clips, but this project acts as a professional director, breaking your idea into six distinct scenes with specific camera angles and visual metaphors. It forces you to approve the full storyboard before generating a single frame, ensuring your final video actually tells a coherent story. It is a fantastic resource for creators who want high-quality, dynamic short-form content without spending hours on manual animation. If you are making educational or motivational shorts, this tool is the shortcut to a polished, professional result.

🆔 @hackernewsgithubprojects

📦 nicolastinkl/hermes_weatherbot

WeatherBot: Autonomous Trading with Hermes

A small Python script now places real bets on Polymarket temperature markets by finding where official weather forecasts disagree with the crowd. Named Hermes, this autonomous bot scans six major US cities every hour to spot mispriced weather buckets. It calculates the true odds using a statistical model and automatically trades when the value looks favorable. The coolest part is that it learns from every trade, adjusting its own risk limits based on past wins and losses. You can watch it all from a text message on your phone. It is a fascinating look at how simple code can handle complex financial decisions.

🆔 @hackernewsgithubprojects

📦 vibe-stack/super-terrain

Sculpt Mountains in Your Browser with Super Terrain

Sculpting a four-kilometer mountain range right inside your browser sounds impossible, but Super Terrain makes it happen. Built on web graphics standards, this tool lets you raise peaks, carve tunnels, and paint surfaces without touching a heavy game engine. It is not just a simple height map; it handles real mesh topology so you can create overhangs and complex shapes. The coolest part is the instant feedback loop. Every brush stroke compiles in the background while the existing scene stays visible, ensuring the heavy lifting never blocks your view. You get professional terrain editing without the usual desktop bloat.

🆔 @hackernewsgithubprojects

📦 openai/privacy-filter

OpenAI Privacy Filter: Instant PII Redaction

Scrub sensitive data from massive text files in milliseconds. OpenAI Privacy Filter is a lightweight model that spots and masks personal identifiers like names, emails, and dates without sending your data to the cloud. The coolest part is how fast it actually is. Instead of guessing word by word, it labels your entire document in one single pass, which means you can run it right on your laptop or even in a web browser. It is fully open source and lets you fine tune it to match your specific privacy rules. This is your fast, local way to sanitize logs before they leak.

🆔 @hackernewsgithubprojects

📦 luckyzyx/luckytool

LuckyTool: Unlocking ColorOS

One specific hack allows LuckyTool to let any app run in split-screen mode, even if the manufacturer explicitly blocked it. This free module for OnePlus phones uses the Xposed framework to quietly change how the operating system works without needing to install a new launcher. It removes annoying ads, lifts restrictions on file names, and gives you control over battery behavior. The project is interesting because it exposes hidden settings that are usually locked away, letting you tailor your device to exactly how you like it. It is a free tool that respects your ownership of the phone.

🆔 @hackernewsgithubprojects

📦 netx-lab/frontier

Frontier

Frontier lets you predict exactly how fast a large language model server will run without spending a single dollar on GPU electricity. This discrete-event simulator replays real traffic patterns through a virtual data center, calculating precise latency for complex features like speculative decoding and sparse models. Researchers use it to test expensive architectural changes on a standard laptop, finding the perfect balance between cost and speed before deploying anything. You get production-level accuracy without the hardware bill. Stop guessing your serving config and start simulating the exact system you plan to build.

📰 https://news.ycombinator.com/item?id=48487049

🆔 @hackernewsgithubprojects

📦 boyang-hu/website-rebuild-skill

How To Rebuild Any Website With AI

A computer agent can now look at a complex website and recreate it so perfectly that the two versions look exactly the same to the human eye. This tool, called website-rebuild-skill, takes the messy, compressed code that powers a live site and untangles it into clean, readable instructions. It is not just a copy; it is a forensic reconstruction that proves every button and animation works identically. This is useful because it turns a black box website into a clear, editable blueprint. The takeaway is simple: you can now learn exactly how a site is built by letting AI reverse engineer it for you.

🆔 @hackernewsgithubprojects

📦 vercel-labs/portless

Portless: Stop Guessing Port Numbers

Local development usually forces you to memorize which random port your app is running on, a frustrating habit that breaks when you switch between multiple projects. Portless eliminates this guesswork by replacing those shifting numbers with stable, named local addresses. Once installed, you simply run your usual dev command through Portless, and it handles the rest by assigning a random port in the background while presenting your app at a clean, predictable URL like myapp.localhost. This works over secure HTTPS by default, meaning hot reloads and cookies behave exactly like they would in production without any extra configuration.

🆔 @hackernewsgithubprojects

📦 gangtailorupgrade/coomertool

CoomerTool: The Smart Archiver

CoomerTool can juggle up to two hundred and fifty-six simultaneous downloads while keeping your file system perfectly organized. It is a Python utility built specifically for archiving media from Kemono and Coomer. The real magic is how it handles interruptions. If your connection drops, it remembers exactly where it stopped and picks up from that exact byte, so you never restart a large download from scratch. It even uses a database to track what you have already grabbed, ensuring you never waste space downloading duplicates. It organizes everything into clean folders and saves the post details right alongside the files.

🆔 @hackernewsgithubprojects

📦 soirihiroka/shrimply

Shrimply: The AI-Powered Video Editor

A free, open-source video editor built in Rust is quietly becoming a powerhouse for creators who want more than just simple cuts. Shrimply brings the flexibility of a full non-linear editor to your desktop, featuring a modern interface and a timeline that handles complex audio and video tracks with precision. The real surprise is its deep integration with artificial intelligence. It includes built-in tools for automatic transcription, text-to-speech, and even AI-driven video generation, all running locally on your own machine. This means you can generate new footage or sync voices to your animations without ever leaving the app or sending your data to a cloud server.

🆔 @hackernewsgithubprojects

📦 vincentwei1021/video-talkcraft

Video Talkcraft

Video Talkcraft is the AI agent skill that turns your coding assistant into a motion design studio for narrated videos. You simply give it a script and a voice recording, and it handles the heavy lifting. It maps every single word to a precise moment in time, ensuring that every visual animation hits exactly when you say it. The coolest part is its anti-slideshow system, which forces constant motion so your video never feels like a static PowerPoint presentation. It handles everything from camera moves to sound effects automatically. You get a professional, polished explainer video without touching a single timeline.

🆔 @hackernewsgithubprojects

📦 googlechromelabs/webmcp-tools

WebMCP Tools: Let AI Drive Your Website

WebMCP Tools is the developer toolkit that lets AI agents drive your website like a human user. Google Chrome Labs built this collection to make the WebMCP API practical, providing utilities and demos that show exactly how to connect language models to web interfaces. The most surprising demo here is a maze game where you play entirely by typing prompts to an AI, with no keyboard or mouse input at all. This shows how a web page can expose structured actions to an agent, replacing fragile screen scraping with direct, reliable control.

🆔 @hackernewsgithubprojects

📦 ezzy1630/downright

Downright: The Native Markdown Editor for Mac

Turn plain text files into beautiful, native Mac documents. Downright is a free, open source editor for macOS that renders Markdown with crisp, system-level typography. It solves the problem of clunky web-based editors by working entirely offline with your local files. The most surprising part is how it handles your text. It keeps your original source code exactly as you wrote it while displaying a polished preview. No hidden formats, no lock-in. It feels like a standard Mac app, not a browser window. If you write notes or docs on a Mac, this tool keeps your files simple and your workflow clean.

🆔 @hackernewsgithubprojects

📦 miaai-lab/qwen3.8-27b-dflash2-exl3-5.0bpw

Qwen 3.8 27B DFlash2 EXL3 5.0bpw

A twenty-seven billion parameter AI model now runs on a single consumer graphics card while generating text almost thirty percent faster than standard setups. This repository provides a complete deployment kit for the Qwen 3.8 27B model, specifically optimized to fit inside twenty-four gigabytes of GPU memory. It solves the problem of running massive, high-quality language models on local hardware without sacrificing speed. The system uses a specialized quantization format to shrink the model size and employs a smaller companion model to guess words, allowing the main engine to verify them rapidly. This makes it possible to run complex reasoning tasks on your own desktop.

🆔 @hackernewsgithubprojects

📦 kingoftheseas56/colosseum

Colosseum: One App for All Your Media

Colosseum is the native desktop application that finally unifies your entire media library. Instead of juggling separate apps for manga, audiobooks, and movies, it brings them all into one fullscreen home. The most surprising part is how it handles your own files. A feature called Vault scans your local folders to index your existing movies and comics. It then opens them directly in the built-in readers and player, so your personal collection lives right next to your downloaded media. It even remembers your spot across every medium. If you have ever wanted one clean place for all your stories, this is it.

🆔 @hackernewsgithubprojects