📦 bymedion/macos-wine-steam

Run Windows Steam Games on Apple Silicon Mac Easily

Merlot is the one-click solution that simplifies playing Windows Steam games on your Apple Silicon Mac. By automating the complex setup of Wine and DXMT, it handles the entire installation process, including downloading the necessary components and configuring the environment for optimal performance. You can use its dedicated app launchers to manage specific game presets directly from your Applications folder or use the generic launcher for a broader setup. It streamlines the technical heavy lifting, allowing you to focus on your games instead of managing backend compatibility layers, providing a smooth path to enjoying your library on macOS.

🆔 @hackernewsgithubprojects

📦 sam1am/backlogia

Organize Your Entire Gaming Library with Backlogia

Consolidate your fragmented digital gaming collection into one unified library by using Backlogia. This local application aggregates games from major platforms like Steam, Epic, GOG, and Xbox into a single interface with smart deduplication. It enriches your collection with high-quality metadata, ratings, and screenshots from IGDB, while providing advanced filtering and sorting options to help you manage your backlog. You can create custom themed collections and even discover new titles based on your unique play history. By running entirely on your own machine, Backlogia keeps your library private and perfectly organized, making it easy to decide what to play next.

🆔 @hackernewsgithubprojects

📦 marcellm01/tinysearch

Local LLM Web Research with TinySearch

TinySearch empowers local large language models by providing a lightweight, source-grounded web research engine that functions entirely through local infrastructure. It replaces bulky search setups by automating the process of performing web searches, reranking the most relevant results, and intelligently crawling pages to extract key data. By transforming raw web content into structured prompts, it ensures that your AI agents receive concise, evidence-based context while keeping citation URLs attached to every piece of information. This tool is designed for developers who want to give their local agents efficient, reliable research capabilities without needing enterprise-scale systems.

📰 https://news.ycombinator.com/item?id=48153340

🆔 @hackernewsgithubprojects

📦 arcosoph/nanowakeword

Build Custom Wake Word Models Easily with Nanowakeword

Nanowakeword provides a complete end-to-end ecosystem for training high-accuracy, custom wake word models that deploy seamlessly from tiny edge devices to powerful cloud servers. This framework simplifies the entire lifecycle by automating data engineering, including phonetic adversarial negative generation to reduce false positives, and advanced model training using techniques like importance sampling based on loss. With support for diverse architectures—ranging from lightweight DNNs to state-of-the-art Transformers and Conformers—it allows for precise optimization of speed and noise robustness. This tool makes creating responsive, reliable voice triggers accessible for any project while keeping deployment efficient and flexible.

📰 https://news.ycombinator.com/item?id=47838289

🆔 @hackernewsgithubprojects

📦 bigstationw/comfyui-untwisting-rope

Master Training-Free Style Transfer with Untwisting RoPE for ComfyUI

ComfyUi-Untwisting-RoPE is the custom node suite that enables high-quality, training-free style transfer for DiT models directly within your existing workflow. By patching model attention layers and rescaling rotational position embedding frequencies, this tool allows you to accurately transfer styles between images without the need for additional model training. It includes powerful features like RF Inversion to maintain noise-matched trajectories and various gating mechanisms to refine the stylistic output. Whether you are using Flux or Z-Image models, these nodes provide precise control over how your target images inherit style from reference inputs, giving you professional results instantly.

🆔 @hackernewsgithubprojects

📦 plantpotatoonmoon/hpsv3-plusplus

HPSv3++: Scaling Reward Models for AI Image Generation

HPSv3++ is the advanced reward model that enables consistent image preference scoring across diverse generative capabilities and reinforcement learning stages. This repository provides a two-stage training pipeline and evaluation framework for a reward model that uses a capability encoder and feature-wise linear modulation to adapt to different generator qualities. It addresses the challenge of maintaining accurate preference signals throughout the optimization process of diffusion models. Users can utilize the provided scripts and weights to score images based on text-following and aesthetic quality, ensuring reliable feedback during development. Incorporate this tool to refine your AI image generation workflows.

🆔 @hackernewsgithubprojects

📦 wanhaoniu/eagg

Generate Precise Robotic Grasps Across Different Hands

Create object-conditioned grasps for diverse robotic hands and grippers by leveraging their specific geometry and structure. This project solves the challenge of generalizing across different hand embodiments by using topology-aware graphs and iterative geometry injection to keep the generator synchronized with evolving contact points during sampling. By processing articulated states into geometry-aware tokens, the system enables effective grasping without forcing heterogeneous hardware into a shared control space. You can use these tools to generate and visualize high-quality grasp candidates across various end effectors and objects, providing a robust solution for complex robotic manipulation tasks.

🆔 @hackernewsgithubprojects

📦 kodiqa-solutions/vaults3

Run a Powerful S3-Compatible Storage Server in Under 80MB

Deploy a lightweight, S3-compatible object storage server that runs as a single binary and uses less than 80MB of RAM. This tool provides a built-in React web dashboard for managing files, audit trails, and bucket policies without needing heavy external dependencies. It supports advanced features like AES-256 encryption at rest, bucket versioning, object locking, and even full-text search across your stored metadata. By combining high performance with a low memory footprint, it offers a robust alternative to larger storage solutions for your infrastructure needs. Simplify your object management and start serving your data efficiently today.

🆔 @hackernewsgithubprojects

📦 jamesmontemagno/pdf-diff

Compare PDF Files Securely and Privately

Compare your PDF documents directly in your browser with complete privacy using this tool. Because it processes files entirely on your device, your data never leaves your computer or gets uploaded to a server, ensuring total security. The interface offers multiple comparison modes including side-by-side, unified, and filtered views to highlight additions or removals instantly. You can even generate formatted PDF reports from your results or use the command-line interface for automated tasks. It is a highly efficient way to spot document changes without risking your sensitive information, giving you peace of mind while staying productive.

🆔 @hackernewsgithubprojects

📦 dariulone/cyberpunk-vr-port

Play Cyberpunk 2077 in Full VR

Transform Cyberpunk 2077 into a fully immersive virtual reality experience by injecting OpenXR head tracking and stereo rendering directly into the game engine. This project adds a complete full-body avatar with motion-controlled hands, enabling you to aim weapons independently and perform motion-based melee attacks. You can manage your gear with immersive hand-to-holster equipping and customize everything through a convenient in-headset overlay. By merging VR controller inputs with native game mechanics, this mod bridges the gap between traditional gameplay and virtual reality. Download this experimental port to bring the streets of Night City to life in your headset.

🆔 @hackernewsgithubprojects

📦 1999kevin/awesome_medical_world_models

Understanding Medical World Models: A Curated Research Collection

Navigate the complex intersection of artificial intelligence and healthcare by exploring a focused collection of academic papers dedicated to medical world models. This repository centralizes research on representing medical states, modeling clinical dynamics, and guiding intervention policies across diverse modalities like X-rays, MRI scans, and electronic health records. By documenting techniques such as latent predictive architectures and spatiotemporal trajectory modeling, the project serves as a key resource for researchers building patient or organ-level simulators. Delve into this curated library to understand how world models are advancing medical diagnosis and clinical decision support systems for more informed patient care.

🆔 @hackernewsgithubprojects

📦 vercel-labs/skills

Master Your Coding Agents with This CLI Tool

Extend the capabilities of your favorite coding agents by managing reusable instruction sets directly from your terminal. This tool provides a unified command line interface to discover, install, and update agent skills, which are structured instruction files that help your AI perform specialized tasks like generating release notes or integrating external workflows. By supporting dozens of popular agents, it simplifies how you maintain and share custom behaviors across your development projects. Whether you are adding skills globally or locally, this utility ensures your coding assistants stay powerful and consistently configured, letting you focus on writing great software.

🆔 @hackernewsgithubprojects

📦 ekkolearnai/hermes-studio

Manage Your AI Agents with Hermes Studio

Control your AI experience by using this comprehensive web dashboard designed for the Hermes Agent. It allows you to manage multi-platform chat sessions, monitor usage analytics, and configure scheduled jobs all from one centralized interface. You can handle profile-specific file uploads, run coding agents, and even interact with voice input or output across various platforms like Telegram, Discord, and Slack. With support for local SQLite storage and robust session management, it makes organizing your AI workflows simple and efficient. Whether you are using the desktop app or a containerized setup, take full command of your agent runtime today.

🆔 @hackernewsgithubprojects

📦 gglucass/headroom-desktop

Slash Your Claude Code and Codex Token Costs by 50%

Reduce your AI token costs by half without changing your coding workflow by using this local-first desktop application. Designed for macOS, the app routes your Claude Code and Codex traffic through an automated optimization pipeline that reversibly compresses prompt bloat, logs, and tool outputs before they ever reach the LLM. It manages its own isolated Python runtime and provides a clear dashboard to track your real-time token savings and pipeline health. By stripping out redundant data while keeping everything retrievable on demand, this tool ensures your existing AI plan stretches twice as far, giving you more efficient development cycles today.

🆔 @hackernewsgithubprojects

📦 dtc111111/gaussiandwm

GaussianDWM: A Unified World Model for Autonomous Driving

GaussianDWM serves as a unified framework that bridges scene understanding and multi-modal generation for autonomous driving. By utilizing 3D Gaussian Splatting, it effectively encodes complex driving environments into a representation that language models can process. This capability allows the system to perform accurate task-aware sampling, which solves the challenge of handling dense 3D information while preserving vital spatial details for future scene prediction. It supports both detailed question answering and dual-condition generation, providing a versatile tool for analyzing driving scenarios. This unified approach enables researchers to bridge the gap between simple comprehension and dynamic, coherent spatial scene synthesis.

🆔 @hackernewsgithubprojects

📦 bikini/exploitarium

Exploitarium: A Collection of Public Exploit Proofs

This repository functions as a consolidated archive of public proof-of-concept exploits and vulnerability research writeups covering a wide range of software. It serves as a centralized resource where each folder contains self-contained research, including implementation details for various vulnerabilities in applications like 7-Zip, AnyDesk, and c-ares. The project includes code to verify these findings, such as Python scripts for reproducing COM impersonation or C harnesses that demonstrate heap-sensitive control-flow manipulation. By documenting these research paths and providing functional proofs, this archive allows developers and security researchers to audit vulnerabilities and understand the technical mechanics behind current exploits.

🆔 @hackernewsgithubprojects

📦 hamed-elfayome/claude-usage-tracker

Track Your Claude AI Usage on macOS

Monitor your Claude AI usage limits in real-time directly from your macOS menu bar with this native application. Built using Swift and SwiftUI, the project provides a seamless way to keep an eye on your consumption and stay informed about your remaining session capacity. It features a detailed six-tier pace system to visualize your usage urgency, customizable color modes for the interface, and clear labels for status tracking. By integrating directly into your system menu bar, it removes the need to manually check limits, helping you manage your AI workflow more efficiently and effectively throughout the day.

🆔 @hackernewsgithubprojects

📦 quant-sentiment-ai/claude-equity-research

Generate Institutional-Grade Equity Research with Claude

Generate professional equity research reports directly within your workflow by installing this plugin for Claude Code. This tool pulls real-time market data, earnings reports, and analyst coverage to produce comprehensive analysis, including fundamental metrics, technical indicators, and risk assessments. You can quickly run institutional-style evaluations for any ticker with actionable buy, sell, or hold recommendations that incorporate probability-weighted scenarios. It is an excellent resource for students or analysts looking to automate preliminary due diligence, providing structured insights for US-listed stocks. Simply use the slash command to start analyzing today, and always remember to conduct your own independent research.

🆔 @hackernewsgithubprojects

📦 luwill/research-skills

Automate Academic Research with AI Agent Skills

Research workflows are being transformed by a collection of specialized AI agent skills designed to handle academic tasks with high precision. This project encapsulates common research processes into portable modules that integrate directly into Claude Code. It offers tools for writing systematic medical imaging literature reviews, generating professional academic slide decks from PDFs with automatic figure extraction, and drafting PhD-level research proposals. A collaborative multi-agent system also allows five distinct roles to work in tandem on survey papers. By adopting these structured workflows, you can replace manual drudgery with a more reliable and consistent research process.

🆔 @hackernewsgithubprojects

📦 jane-xiaoer/xiaoer-videolab

Xiaoer VideoLab: One-Click Local Video Downloads

Xiaoer VideoLab is the browser extension that finally lets you download videos directly to your local folder with a single click. Instead of relying on privacy-invasive extensions, this project uses a tiny, local background daemon powered by the battle-tested yt-dlp to handle your downloads. It supports over 1800 sites, including YouTube, Bilibili, and Instagram, without sending your data to external servers. Because all the processing happens right on your machine, you get full control over your files and keep your browsing private. Install the engine once and simplify your entire video-saving workflow for good.

🆔 @hackernewsgithubprojects

📦 saihgupr/dmgmaker

Create Professional macOS Installers with DMGMaker

Package your macOS applications into professional disk images using this dedicated creation tool. It simplifies the process by generating high-quality installers featuring live-rendered SwiftUI mesh gradients and a sleek glassmorphic interface that looks sharp on Retina displays. You can drag and drop your files directly into the intuitive interface or utilize the command-line support to integrate DMG creation into your automated build pipelines. By including specialized naming logic to prevent Finder display issues, this project ensures your installer looks polished and professional. Grab this tool to streamline your software distribution and create refined, user-ready installers effortlessly.

🆔 @hackernewsgithubprojects