📦 leookun/cursor-byok

cursor-byok

The cursor-byok project is the local companion app that finally lets you use your own custom API keys and models inside the Cursor editor. If you love the editor but want complete control over your AI costs and privacy, this project solves that by routing your editor requests through a local proxy server. It acts as a bridge, letting you swap out the default AI backend with your own custom endpoints and API keys. You get to run whatever models you want, keep your code local, and customize your setup using a clean local interface. It is the perfect way to make your favorite coding tool truly yours.

🆔 @hackernewsgithubprojects

📦 carsteneu/ai-memory-comparison

ai-memory-comparison

Cut through the marketing noise and compare dozens of memory systems for AI coding agents with zero bias. This open-source repository, called ai-memory-comparison, evaluates dozens of different platforms across nearly eighty distinct features to help you find the absolute best way to give your digital assistant a long-term memory. Instead of relying on empty promises, every single feature rating in this table is backed by a direct link to actual public source code or documentation. It is the ultimate factual guide for anyone trying to build smarter, context-aware coding tools without getting lost in the hype.

🆔 @hackernewsgithubprojects

📦 beefiker/superloopy

Superloopy

Superloopy is the lightweight loop harness for Codex and Claude Code that finally forces AI agents to back up their work with real, verified proof. Instead of just letting an AI confidently claim a task is done, you prefix your command with loopy, and the system immediately builds a plan and forces the agent to generate actual, verifiable file artifacts. It even reruns your test commands at the end to make sure nothing was faked or broken along the way. It is a brilliant way to keep coding agents honest and accurate. Check out the project to start running bulletproof AI development loops today.

🆔 @hackernewsgithubprojects

📦 anagraphy/roampage

Roampage

Organize your self-hosted services and homelab apps into a beautiful, responsive dashboard without writing a single line of configuration code. This clever project lets you add, edit, and reorder your categories and servers directly from your web browser using a smooth visual interface. Instead of dealing with messy configuration files, you get instant inline editing, live health checks for your services, a library of over twenty-five hundred icons, and custom tags with a built-in color picker. You can customize the look of every page with your own wallpapers and custom CSS styles that apply instantly. It runs in a simple Docker container, making it the perfect elegant hub for your...

🆔 @hackernewsgithubprojects

📦 whiskeycoder/qwen3-audiobook-converter

qwen3-audiobook-converter

Turn your digital library into realistic audiobooks using a powerful, open-source voice synthesizer running right on your own machine. This clever command-line tool takes your local files like PDFs, EPUBs, or Word documents and pipes them directly through a local voice model to generate natural-sounding narration. The coolest part is the voice cloning feature, which lets you feed in a short audio sample of any voice to clone it automatically without needing to transcribe a single word manually. It handles long books by smart-chunking the text, caching progress so you never lose your spot, and outputs a complete, beautifully stitched MP3. Grab your favorite e-books and start building your custom...

🆔 @hackernewsgithubprojects

📦 modelscope/awesome-vibe-research

Awesome Vibe Research

Supercharge your academic workflow by exploring a massive, community-driven collection of artificial intelligence tools built specifically for scientific discovery. The awesome-vibe-research repository serves as a central hub that organizes open-source agents, custom skills, and automated workflows across every single stage of the scientific process. Instead of getting bogged down in repetitive tasks, you can find specialized systems that help with everything from generating initial hypotheses and running unattended experiment code to drafting paper sections and creating high-quality figures. It is an incredible map of practical resources that makes modern, AI-assisted research accessible to any curious mind.

🆔 @hackernewsgithubprojects

📦 philfung/open-mrna

Open mRNA

Open-mRNA is the open-source blueprint that finally lays out how to design and synthesize a personalized cancer vaccine in a private laboratory setting. It maps out an incredible end-to-end journey starting with a patient's tumor biopsy, running the raw genetic data through advanced neural networks to identify custom target mutations, and translating those digital designs back into a physical, highly stable vaccine formulation. By bridging the gap between clinical bioinformatics and benchtop laboratory equipment, this project makes cutting-edge vaccine design understandable for anyone interested in the future of medicine. Check it out to see how the code is written.

🆔 @hackernewsgithubprojects

📦 srilaxmi1993/generative-ai-projects

Generative AI Projects

You can now learn to build practical AI agents without getting bogged down in dense machine learning theory. The generative-ai-projects repository is a hands-on, beginner-friendly guide packed with runnable code for single and multi-agent systems. Instead of just reading about AI, you get to build real applications like customer support routers, automated market researchers, and cooperative multi-agent travel planners. It uses popular tools like LangChain, CrewAI, and AutoGen to show you how real-world systems manage state, search documents, and talk to APIs. It is the perfect playground to stop theorizing and start building actual AI products today.

🆔 @hackernewsgithubprojects

📦 rebelytics/one-skill-to-rule-them-all

One Skill to Rule Them All

Supercharge your AI workflows by letting your system automatically build and refine its own capabilities while you work. This project introduces task-observer, a brilliant meta-skill that runs quietly in the background during your normal work sessions. Instead of freezing your AI prompts in time, it watches what you do, spots repetitive patterns to draft new custom skills, and captures your manual corrections to update existing ones. It even monitors and improves its own methodology over time. You simply work, and your entire skill library evolves with you. It is the ultimate way to build a self-improving digital assistant that gets smarter every single day.

🆔 @hackernewsgithubprojects

📦 fareedkhan-dev/autonomous-agentic-rag

Autonomous Agentic RAG

Build a search assistant that actually learns how to fix its own mistakes. The autonomous-agentic-rag project moves past rigid, hard-coded setups by letting a team of AI agents continuously rewrite their own instructions. First, specialist agents dig through databases to answer complex queries, while an evaluator scores their work. Then, a diagnostic agent spots the weakest link and an architect agent updates the team's rules to patch the flaw. It basically runs an evolutionary loop to find the best possible strategy, meaning your search assistant gets smarter over time without you writing new code.

🆔 @hackernewsgithubprojects

📦 zzxscodes/trading-system-notes

Trading System Notes

Trading-system-notes is the open-source knowledge base that demystifies how high-frequency trading platforms achieve near-zero delay. Instead of hidden proprietary code, this repository pulls back the curtain on the exact hardware and software tricks used by top Wall Street engineers. It documents how to configure computer memory, align data to modern microchips, and tune operating systems so programs can process market data in nanoseconds. It is incredibly useful because it compiles scattered industry wisdom on low-latency engineering into one clear guide. If you want to understand how computer systems are pushed to their absolute physical speed limits, this notebook is your perfect starting point.

🆔 @hackernewsgithubprojects

📦 kocasariumut/faceanything

Face Anything

Reconstruct detailed, moving 3D face models from any standard video clip or single photo. This system uses a smart approach called canonical point prediction, assigning every pixel a coordinate in a shared 3D space to track facial movements over time. Instead of relying on complex multi-camera setups, it processes standard video frames to generate temporally consistent 3D geometry, camera poses, and colored point clouds. It even automatically isolates faces by removing background clutter. Just feed in a quick clip of someone talking, and you get back an incredibly detailed, rotating 3D portrait showing real-time expressions.

🆔 @hackernewsgithubprojects

📦 nexaflowfrance/openfamily

OpenFamily

OpenFamily lets you run a complete digital command center for your household right from your own home server, keeping your private family data entirely under your own control. It is a self-hosted, open-source alternative to commercial organizing apps that puts your shopping lists, chores, meal plans, shared calendars, and household budgets in one place. What makes it incredibly cool is how easily it plugs into other home-server favorites, letting you automatically sync recipes from Mealie or track your groceries through Home Assistant. It is the perfect, privacy-first way to keep your whole household on the same page.

🆔 @hackernewsgithubprojects

📦 dammyjay93/interface-design

Interface Design

AI coding agents are notorious for building interfaces with inconsistent spacing, drifting button heights, and chaotic color palettes that change every single session. The interface-design repository solves this by acting as a dedicated design engineer for tools like Claude Code and Codex, giving your agent a persistent visual memory. It automatically saves your UI decisions to a local markdown file, ensuring every new page or component strictly matches your established spacing grids, depth strategies, and surface treatments. Instead of starting from scratch every time you prompt, your agent instantly loads your custom system to keep dashboards and admin panels perfectly unified. It is the ultimate way to build clean, professional...

🆔 @hackernewsgithubprojects

📦 autobrr/qui

qui

Managing multiple qBittorrent downloads across different servers usually means keeping a dozen browser tabs open and constantly toggling between them. This is where qui steps in, combining all of your separate download instances into a single, incredibly fast dashboard. Packaged as a lightweight application with absolutely no external dependencies, it lets you control every server from one central place while keeping your interface clean. It even includes clever features like automatically cross-seeding matching files across different trackers and running automated cleanup rules. If you run multiple download setups and want a simpler way to control them, this tool is definitely worth checking out.

🆔 @hackernewsgithubprojects

📦 wh-2099/mermaid-skill

Mermaid Skill for Claude Code

You can now generate twenty-three different types of Mermaid diagrams directly inside Claude Code using simple text prompts. The mermaid-skill repository is a clever extension that integrates with your terminal assistant, allowing you to quickly spin up flowcharts, database designs, or sequence diagrams with simple commands. Instead of fighting with complex syntax, this tool automatically references its own pre-packaged documentation to structure, lay out, and style your diagrams perfectly. It even uses a background process to keep its syntax guides synchronized with the official Mermaid project. It is like having a professional designer in your terminal who instantly turns raw ideas into clean, visual architecture maps.

🆔 @hackernewsgithubprojects

📦 leerob/pixo

Pixo

Compress your images to incredibly small sizes without depending on bloated external libraries. This project brings you a blazing fast image compressor built entirely from scratch in Rust, supporting both PNG and JPEG formats. Instead of wrapping heavy legacy tools, it implements the math and logic itself, allowing you to run it anywhere from a lightweight command line tool to a tiny WebAssembly binary in your web browser. It packs advanced features like palette quantization for lossy PNGs and smart trellis quantization for JPEG, making your files remarkably compact. It is the ultimate tool for developers who want ultra-clean code and tiny bundle sizes.

🆔 @hackernewsgithubprojects

📦 tonyd2wild/deepseek-v4-flash-dspark-1m-nvfp4-kv-2x-dgx-spark

deepseek-v4-flash-dspark-1m-nvfp4-kv-2x-dgx-spark

Run massive AI models with a massive one-million-token context window on just two nodes. This project provides a validated recipe to serve the DeepSeek V4 Flash model with speculative decoding across a pair of high-performance servers. By utilizing a highly compressed attention cache format, it fits nearly two million total tokens into active memory. It specifically solves the problem of model degradation, like endless loops or random text drift, when running high-concurrency agent workflows. Now, developers can easily handle multiple long-context sessions simultaneously without crashing their hardware or sacrificing speed. Give this setup a spin to scale your local AI agents.

🆔 @hackernewsgithubprojects

📦 balukosuri/andrej-karpathy-s-autoresearch-as-a-universal-skill

Andrej Karpathy's Autoresearch as a Universal Skill

Drop this self-improving prompt optimizer into any code repository, and it automatically scans your codebase, suggests what to optimize, defines clear evaluation metrics, and runs an autonomous loop that makes your prompts better over time while you sleep. Inspired by Andrej Karpathy's method of letting an AI agent autonomously run and evaluate code experiments, this universal tool brings that exact evolutionary process to prompt engineering. Instead of manually tweaking prompts based on vibes, the system defines strict binary yes-or-no criteria to grade performance, runs continuous tests, keeps what works, and throws away what doesn't. It is the ultimate hands-off way to build battle-tested prompts for any project.

🆔 @hackernewsgithubprojects

📦 vasilyevdm/ai-agent-handbook

ai-agent-handbook

The ai-agent-handbook is the master engineering reference that finally pulls back the curtain on how thirty of the biggest AI agent frameworks actually work under the hood. Instead of surface-level tutorials, this project did a deep dive into real production codebases like LangGraph and Claude Code to figure out why agents fail and how to fix them. The coolest takeaway is how they handle context rot, showing that agent performance drops when the context window is only a quarter full because bloated tools eat up all the memory. It is a goldmine of real architectural patterns for building smarter, leaner AI systems.

📰 https://news.ycombinator.com/item?id=47461861

🆔 @hackernewsgithubprojects