📦 matheuzsecurity/furtex

furtex

Linux security systems are completely blind to this. The furtex toolkit executes complex post-exploitation and evasion techniques by using raw kernel operations instead of standard system calls, which entirely bypasses traditional endpoint detection. Instead of relying on common system call tables that security agents monitor, this project uses io_uring workqueues and custom BPF map tools to read files, establish network connections, and inject code silently. It is a fascinating playground for researchers trying to understand the next generation of stealth techniques, demonstrating how modern kernel features can be repurposed. You can compile it directly to test how your own defenses hold up against these quiet tactics.

📰 https://news.ycombinator.com/item?id=48999323

🆔 @hackernewsgithubprojects

📦 akitaonrails/ai-memory

AI Memory

AI Memory is the developer tool that keeps your coding agents in sync across different terminal sessions and providers. If you have ever been frustrated by an AI assistant losing all context the second you close the window, this utility solves it by automatically capturing session milestones and writing them to a shared git-backed markdown wiki. You can literally shut down Claude Code mid-task, fire up Codex or Cursor in the exact same workspace, and the new agent immediately receives a clear handoff of where you left off. It is the ultimate persistent brain for your developer agents.

🆔 @hackernewsgithubprojects

📦 marcelroed/gigatoken

Gigatoken

Slash your language model tokenization times from hours to mere seconds by switching to gigatoken. This clever project delivers a drop-in replacement for Hugging Face and tiktoken that runs up to a thousand times faster, letting you process text data at multiple gigabytes per second. Instead of getting bogged down by slow regex engines during pre-tokenization, it uses highly optimized rust code, custom SIMD operations, and a smart caching system to fly through text. It is a game-changer for anyone dealing with massive datasets, allowing you to tokenize billions of words on a standard CPU over your morning coffee break.

🆔 @hackernewsgithubprojects

📦 justvugg/nanoeuler

nanoeuler

You can train a hundred and sixteen million parameter language model on a single graphics card using code written entirely from scratch in C and CUDA. A brilliant developer built nanoeuler to bypass heavy frameworks like PyTorch completely, hand-writing everything from the tokenizers to the actual math equations that run the neural network. By coding the forward and backward training passes by hand, they created an educational masterpiece that pretrains on classics and web text, then fine-tunes into a chat assistant. It is a beautiful, raw demonstration of pure engineering that shows exactly how these massive AI systems compute under the hood.

📰 https://news.ycombinator.com/item?id=48710778

🆔 @hackernewsgithubprojects

📦 postgresql-tools/migra

migra

Compare two PostgreSQL database schemas and instantly generate the SQL migration script needed to transition from one to the other. Instead of manually writing tedious alter table commands, you just point this tool at your production database and your development branch. It compares tables, constraints, indexes, views, and security policies, then outputs the clean SQL to sync them up. If you prefer not to connect to a live database, you can simply feed it schema dump files instead. It also features optional smart AI breakdowns that explain migration risks, generate rollbacks, and advise on potential locking issues before you deploy.

📰 https://news.ycombinator.com/item?id=48328343

🆔 @hackernewsgithubprojects

📦 chenglou/freerange

Freerange

Catch layout bugs and math errors in your TypeScript code at compile time without ever opening a browser. Freerange analyzes ordinary TypeScript code to track the exact ranges of your numbers, instantly flagging potential division by zero, out-of-bounds array indexes, or layout calculations that might result in NaN or Infinity. Instead of introducing a complex new language or messy annotations, it lets you use simple, static console assertions to set up guardrails for your UI code. It is a incredibly fast tool that works directly with the official TypeScript API to guarantee your calculations are completely safe before your app even runs.

📰 https://news.ycombinator.com/item?id=48987176

🆔 @hackernewsgithubprojects

📦 crawlseo/crawlseo

CrawlSEO

CrawlSEO is the self-hosted dashboard that finally makes SEO monitoring painless for founders who do not want to pay hundreds of dollars a month for bloated agency tools. It brings Google Search Console data, Core Web Vitals, and a built-in site crawler together in one free dashboard. You can run concurrent audits on up to two thousand pages, map performance scores, and track keywords with direct delta comparisons. Best of all, it has an integrated Model Context Protocol server, meaning you can hook up AI agents like Claude to query your SEO data or trigger site crawls directly from your developer terminal.

🆔 @hackernewsgithubprojects

📦 sqliteai/adam

Adam

The entire AI agent loop now fits into a single C include file. That is the magic of adam, an embeddable, cross-platform library that acts like the SQLite of AI frameworks. Instead of dragging in massive python environments, this tiny C library compiles anywhere, from iOS to WebAssembly, giving you tool calling, long-term memory, and full audio pipelines. It even runs local models right on your device. You can even embed it as a SQL extension in SQLite or Postgres so your database can literally query itself using natural language. It is a incredibly smart, self-contained way to build AI apps.

📰 https://news.ycombinator.com/item?id=48035677

🆔 @hackernewsgithubprojects

📦 eriktromp/colibri-hy3

Colibri Hy3

Colibri-hy3 is the lightweight inference engine that finally lets you run a massive two hundred and ninety-five billion parameter language model right on a normal consumer computer. Instead of requiring an expensive cluster of enterprise graphics cards, this clever C project keeps only a tiny five-gigabyte core of the model in your active system memory. The rest of the colossal model lives on your solid-state drive, streaming individual network experts directly into your processor on demand. It is a brilliant way to experiment with giant neural networks on a budget, turning a standard laptop into an AI powerhouse.

🆔 @hackernewsgithubprojects

📦 nexu-io/motion-anything

Motion Anything

Animate web pages and build launch videos using nothing but natural language. This repository houses an open-source, chat-native motion engine driven by coding agents. Instead of manually tweaking complex keyframes or writing endless lines of CSS, you can simply type out how you want a transition to feel, and the AI translates that emotion into smooth, real-time code right on the page. It comes packed with hundreds of pre-built motion recipes, design packs, and templates that make UI micro-interactions look professional instantly. It is the perfect tool for making your web designs feel alive without the typical coding headache.

🆔 @hackernewsgithubprojects

📦 rpatrik96/bibtexupdater

BibTeX Updater

Clean up your academic bibliographies and automatically upgrade preprints to their final published versions with bibtexupdater. Writing a paper is hard enough without manually hunting down published versions of the early drafts you cited. This tool scans your bibliography, matches preprints against massive databases like arXiv, Crossref, and OpenAlex, and pulls in the official published metadata while keeping your citation keys intact. Even better, it includes a rigorous fact-checker to detect hallucinated or incorrect references before you submit. It is the perfect way to polish your reference list and keep your research completely accurate.

🆔 @hackernewsgithubprojects

📦 julian-adv/openmmo

OpenMMO

Step into a massive procedural fantasy world where humans and artificial intelligence play together as absolute equals. In this sandbox, AI agents connect using the exact same network protocol as human players, meaning they have no backdoor shortcuts or special treatment. If a human can do it, so can the agents. You can construct multi-story timber houses, survive dangerous real-time battles, and watch the landscape change across a vast world complete with braided river systems and dynamic seasonal day-and-night cycles. It completely redefines multiplayer gaming by creating a living, breathing ecosystem where you cannot tell if your neighbor is a real person or a clever bot.

🆔 @hackernewsgithubprojects

📦 cristibaluta/imagin-raw

Imagin RAW

This lightweight app replaces Adobe Bridge with an active memory footprint of just one hundred megabytes instead of gigabytes of system resources. Built natively for macOS, imagin-raw lets you browse, cull, and organize massive folders of RAW photos instantly without waiting for a slow import process or managing a heavy library database. It decodes RAW camera formats directly on your device, auto-groups matching image pairs, and lets you apply standard ratings and color labels that sync directly with other photo editors. You get a incredibly fast, privacy-focused workspace that launches in milliseconds and keeps your computer running cool while sorting through thousands of photos.

📰 https://news.ycombinator.com/item?id=48995444

🆔 @hackernewsgithubprojects

📦 genseeai/gensee-crate

Gensee Crate

Keep your AI coding agents on a tight leash by monitoring and controlling their system actions in real time. Gensee-crate runs as a low-latency safety guard beside agents like Claude Code and Copilot on macOS and Linux, watching every file they edit, command they run, and network request they make. If an agent tries to read a sensitive credential or write outside your project workspace, the customizable policy engine blocks it before any damage is done. You can easily trace agent behavior across sessions with visual timeline graphs, making it the perfect tool for keeping autonomous AI coding safe and predictable. Secure your local development workspace today.

🆔 @hackernewsgithubprojects

📦 paullililili/sindy4engineers

sindy4engineers

sindy4engineers is the interactive tutorial that finally makes advanced data-driven physics practical for everyday engineering applications. If you have ever wanted to extract the actual mathematical equations governing a complex system just by looking at its sensor data, this project guides you through the entire process from scratch. Instead of relying on complicated, black-box software libraries, it breaks down the core math and code of the sparse identification of nonlinear dynamics method. You will explore real-world engineering scenarios, like discovering the flight physics of a drone or modeling chaotic heat transfer in a fluid system. By the end, you will understand how to turn noisy, real-world data directly into clean...

🆔 @hackernewsgithubprojects

📦 shigella520/mindtrain

mindtrain

The private AI knowledge gym that finally turns temporary chat conversations into permanent memory. MindTrain is a self-hosted platform designed to help you practice and review concepts directly through an AI coach. Instead of just answering questions and moving on, you can pause, ask for hints, or challenge an answer in mid-conversation without ruining your score. The magic happens behind the scenes, where the app uses deterministic grading to keep scores accurate and runs a spaced-repetition scheduler to automatically pause new material if your backlog gets too high. It is the perfect setup for keeping your study data private on your own machine.

🆔 @hackernewsgithubprojects

📦 leeguooooo/chrome-use

chrome-use

Control your real, logged-in Chrome browser directly from any AI agent or command line without ever triggering annoying bot detection or dealing with constant remote debugging popups. While typical automation tools force you into an empty browser where you have to re-login and solve endless captchas, chrome-use connects straight to your active browser session using a lightweight extension and native messaging. This means your cookies and login sessions are already active, and sites read your agent's actions as a real human. You can easily watch the agent work in real-time, step in to solve a puzzle if needed, and let it resume seamlessly.

📰 https://news.ycombinator.com/item?id=48668131

🆔 @hackernewsgithubprojects

📦 experientiallabs/world-model-harness

World Model Harness

AI agents are incredibly powerful, but testing them in real-world environments is expensive, slow, and risky. World-model-harness solves this by letting you build a simulated environment directly from recorded agent traces. Once you record how your agent interacts with tools and databases, this project compiles those actions into a local, lightweight world model that mimics the real system. From there, you can safely run parallel tests and automatically optimize your agent prompts, tools, and code without paying for constant live API calls. It gives developers a safe, sandboxed playground to continuously test and improve their AI agents before deploying them.

📰 https://news.ycombinator.com/item?id=48735124

🆔 @hackernewsgithubprojects

📦 moonshotai/kimi-code

kimi-code

You can now run a fully interactive AI coding agent directly inside your terminal. Developed by Moonshot AI, kimi-code is a command-line tool that reads and edits your project files, runs shell commands, searches code, and surfs the web to solve programming tasks. What makes it special is its real-time feedback loop: it executes changes, tests the results, and automatically decides the next steps. It works out of the box with Kimi models but easily connects to other providers. It is the perfect terminal companion for developers who want an automated assistant to handle the heavy lifting of debugging and coding.

🆔 @hackernewsgithubprojects

📦 1ay1/agentty

Agentty

Agentty is the terminal-based AI pair programmer that starts in under a single millisecond without needing heavy dependencies like Node or Python. It packages a complete, lightning-fast assistant into a single static binary that runs right inside your command line. The absolute coolest part about this project is its entirely local, zero-dependency search engine that runs right on your machine. It combines classic keyword search with smart vector search to instantly find answers across your local files, code, and documentation, even when you are completely offline. You get a private, blazing-fast coding companion that respects your system and respects your data.

📰 https://news.ycombinator.com/item?id=48928002

🆔 @hackernewsgithubprojects