📦 tencent-hunyuan/hunyuan-a13b

Meet Hunyuan-A13B: A High-Efficiency Powerhouse LLM

Are you looking for top-tier AI performance without needing a supercomputer? Meet Hunyuan-A13B, an innovative open-source large language model that solves the trade-off between power and resource consumption. Built on a fine-grained Mixture-of-Experts architecture, it features eighty billion total parameters but only thirteen billion active ones. This design delivers impressive reasoning, coding, and agent capabilities while remaining highly efficient for various environments. It even supports an ultra-long context window of 256K and offers flexible thinking modes for complex tasks. Whether for research or cost-effective AI development, Hunyuan-A13B is a versatile foundation designed to make advanced artificial intelligence more accessible.

🆔 @hackernewsgithubprojects

📦 disler/claude-code-hooks-mastery

Mastering Claude Code Hooks: Ultimate Guide

Ever wanted to take full control over your AI coding assistant? This repository is a masterclass in using Claude Code hooks to customize your workflow. It shows you how to implement deterministic control, like blocking dangerous commands, validating prompts, or even triggering voice notifications. By using Python scripts managed with UV, you can securely intercept tool usage and monitor your AI agent lifecycle in real time. It even features advanced patterns like team-based validation and custom slash commands to streamline your development. Dive in to transform your coding environment into a powerful, automated, and truly personalized AI workspace.

🆔 @hackernewsgithubprojects

📦 wquguru/harness-books

Master AI Agent Design with Harness Engineering

Ever wonder how to keep AI coding agents from going off the rails? Harness engineering is the secret to building stable, accountable systems. This repository offers two comprehensive books that dive deep into the design philosophies behind tools like Claude Code and Codex. Instead of focusing on basic prompts, these books explore essential structural components like query loops, permission systems, and multi-agent verification that turn unstable models into reliable, professional workflows. Whether you are building your own harness or just want to understand the architecture behind top-tier coding agents, these guides provide the frameworks you need to succeed.

🆔 @hackernewsgithubprojects

📦 viperrcrypto/siftly

Turn Your Messy Twitter Bookmarks Into An AI-Powered Library

Are your Twitter bookmarks a disorganized mess you never look at again? Siftly is a self-hosted tool that transforms your bookmarks into a searchable, visual knowledge base directly on your machine. It runs a powerful four-stage AI pipeline, performing entity extraction, vision analysis on media, and semantic tagging to categorize everything for you. You can explore your collection through an interactive mindmap or use natural language to find exactly what you need. It even supports Claude CLI for automatic authentication, keeping your data private and local. Take control of your saved content and make it useful again today.

🆔 @hackernewsgithubprojects

📦 shpigford/chops

Stop Managing Your AI Agent Skills Manually!

Are you tired of digging through hidden system files just to manage your AI agent skills? Meet Chops, a native macOS application designed to finally bring order to your coding workflow. Chops acts as a central hub where you can discover, edit, and organize skills and agents across tools like Claude Code, Cursor, and Windsurf. With features like real-time file watching, full-text search, and a built-in markdown editor, it replaces tedious manual configuration with a smooth, native interface. Stop manually editing dotfiles and start managing your AI agents effectively today. It is the perfect way to streamline your development process.

🆔 @hackernewsgithubprojects

📦 fcakyon/claude-codex-settings

Supercharge Your AI Coding Workflow

Want to take your AI coding assistant to the next level? This repository offers a massive collection of battle-tested settings, specialized plugins, and custom agents designed to supercharge tools like Claude Code and OpenAI Codex. It solves the problem of manual setup by providing organized configurations for cloud infrastructure, database management, and even office document automation. Whether you are deploying to the cloud, managing GitHub pull requests, or optimizing complex database queries, these structured skills streamline your entire development process. Dive into this professional setup to automate repetitive tasks and make your AI assistant work significantly harder for you.

🆔 @hackernewsgithubprojects

📦 mage0535/hermes-memory-installer

Stop Your AI Agent From Forgetting Everything!

Are you tired of your AI agent acting like a stranger every time you start a new session? Most coding assistants suffer from total amnesia once a conversation ends, forcing you to repeat your preferences and project constraints endlessly. This repository provides a clever memory sidecar that solves this by running alongside your agent, giving it persistent, production-grade recall without ever touching its internal code. It uses a three-tier architecture—hot, warm, and cold—to archive sessions, track recurring topics in dedicated dossiers, and inject relevant context, ensuring your agent actually learns and gets smarter the more you use it.

🆔 @hackernewsgithubprojects

📦 ethanweber/posterskill

Create Professional Academic Posters Instantly with AI

Struggling to design the perfect academic poster? This tool is a game changer for researchers. By connecting your existing paper and project website, it automatically generates a print-ready poster as a self-contained HTML file. The best part is the live visual editor that lets you drag dividers, swap sections, and adjust font sizes directly in your browser. You can even export your custom layout back to the AI for further refinement, ensuring your final design is polished and professional. Stop fighting with complex layout software and start building beautiful, high-resolution posters for your next big conference today.

🆔 @hackernewsgithubprojects

📦 the-17/agentsecrets

Stop Exposing Your Secrets to AI Agents

Are you tired of worrying about your AI agents leaking sensitive credentials? Standard methods let agents store and handle raw keys, leaving them wide open to theft through prompt injection or malicious plugins. AgentSecrets changes the game with a zero-knowledge infrastructure that completely decouples your credentials from the agent runtime. Instead of holding raw values, your agents operate by reference. Using powerful features like transport-layer injection and process-level capability bounding, this tool ensures sensitive keys never enter your agent memory. Keep your credentials secure while letting your agents do the work, and start building truly trustless AI workflows today.

🆔 @hackernewsgithubprojects

📦 withkynam/vibecode-pro-max-kit

Stop AI Forgetting Your Project Context

Is your AI assistant constantly losing track of your project goals? This repository offers a sophisticated spec-driven coding harness designed to act as your team's persistent memory. It organizes development into a structured system using fifteen specialized agents and over thirty skills to manage everything from research and planning to execution and validation. By implementing the rigorous RIPER-5 methodology, it enforces protocol compliance and kills context rot, ensuring you ship clean features instead of spaghetti code. Whether you are a solo builder or lead a team, this kit keeps your development on track and fully automated. Start building smarter today.

🆔 @hackernewsgithubprojects

📦 weichenfan/uae

Unlock Better Generative Modeling with Unified Autoencoding

Are you looking to bridge the gap between pixel-perfect images and deep semantic understanding? This repository introduces an innovative approach called Unified Autoencoding that harmonizes these two worlds to improve generative modeling. By using a specialized pipeline, it allows for flexible image reconstruction and high-fidelity generation without needing to add extra tokens. Whether you are working in latent space or pixel space, this framework provides the tools to train and evaluate powerful models effectively. Dive into this repository to see how modern autoencoding techniques are reshaping visual data processing and start building your own high-quality generative pipelines today.

🆔 @hackernewsgithubprojects

📦 claudiodrews/memory-os

Stop Your AI Agent from Forgetting Everything

Tired of your AI agent having amnesia every time you start a new conversation? Memory OS solves this by adding a permanent seven-layer memory system to your Hermes Agent. It runs entirely on your local machine, utilizing tools like Qdrant for vector storage, structured fact tracking, and a self-curating wiki to keep your project context sharp. Instead of repeating yourself, the agent uses intelligent, surgical context injection to remember your decisions and preferences across every session. It even enforces a ground truth hierarchy so the agent actually uses what it remembers. Give your AI a reliable long-term memory today.

🆔 @hackernewsgithubprojects

📦 finsys/dockhand

Take Control of Your Docker Containers with Dockhand

Struggling to manage your Docker containers efficiently? Meet Dockhand, a powerful, user-friendly management interface designed to simplify your container workflows. This tool lets you effortlessly monitor your environments, handle batch operations, and track container updates with ease. Beyond simple controls, Dockhand includes a robust logging viewer, file management capabilities, and even built-in vulnerability scanning to keep your systems secure. Whether you are handling complex stacks or just need a cleaner dashboard for your services, it provides the visibility and control you need. Say goodbye to manual CLI headaches and upgrade your container management experience today.

🆔 @hackernewsgithubprojects

📦 waynesutton/opensync

Take Control of Your AI Coding Sessions with OpenSync

Are you struggling to keep track of your AI coding sessions across multiple tools? Meet OpenSync, a powerful dashboard solution designed to sync and organize your activity in one central place. It works seamlessly with popular tools like Claude Code, OpenCode, and Cursor, allowing you to track tool usage, monitor token spend, and manage your history effortlessly. You can even use semantic search to find specific context or label sessions to export as datasets for evaluations. It is the perfect way to gain visibility into your AI workflows and keep your data private. Start syncing your sessions today.

🆔 @hackernewsgithubprojects

📦 zhongkaifu/tensorsharp

Run LLMs Locally with C# & GPU Power!

Want to run large language models right on your own machine? TensorSharp is a powerful C# inference engine designed for local AI performance using GGUF model files. Whether you are on Windows, macOS, or Linux, it leverages full GPU acceleration through backends like CUDA and Metal to keep your models running fast. You get a versatile console tool, an interactive chat REPL, and a web interface that is fully compatible with OpenAI and Ollama APIs. With built-in support for multimodal inputs and advanced batching, it is an incredible way to take control of your AI workflow. Check out TensorSharp today!

📰 https://news.ycombinator.com/item?id=48519136

🆔 @hackernewsgithubprojects

📦 zhenghao2519/spacedrive

SpaceDrive: Giving Autonomous Cars Spatial Awareness

Did you know that standard vision models struggle to understand 3D space in autonomous driving? Enter SpaceDrive, a groundbreaking framework that upgrades vision language models with true spatial awareness. By treating spatial information as explicit positional encodings rather than simple text, it enables models to reason about 3D coordinates more effectively. It integrates multi-view depth and historical data, replacing clunky digit-by-digit generation with precise trajectory regression for better navigation. This approach significantly boosts planning accuracy in real-world environments, pushing the boundaries of what autonomous systems can achieve. Check out this project to see how spatial intelligence is transforming the future of driving.

🆔 @hackernewsgithubprojects

📦 kellyclaudeai/openclaw-deck

Supercharge Your Workflow with OpenClaw Deck

Want to talk to multiple AI agents at the exact same time? Meet OpenClaw Deck, a powerful multi-column chat interface built for your local agents. It solves the clutter of single-chat windows by letting you manage up to seven agent conversations side-by-side in one responsive view. With real-time WebSocket connections, smooth markdown rendering, and eleven beautiful dark themes like Dracula and Nord, it is a productivity powerhouse for developers. You can even use keyboard shortcuts to jump between columns instantly. It is an amazing way to keep your workflow organized and efficient. Give it a try and transform how you chat!

🆔 @hackernewsgithubprojects

📦 patientx-cfz/comfyui-rocm

Boost Your AMD GPU Performance in ComfyUI

Are you struggling to get top-tier AI performance on your Windows machine with an AMD graphics card? This repository is a dedicated solution designed to bridge that gap. By leveraging AMD’s official ROCm and PyTorch libraries, it optimizes your setup specifically for better performance. The project includes a seamless auto-installation process and comes pre-packed with popular performance-enhancing tools like triton, flash-attention, and bitsandbytes. It is a fantastic way to ensure your hardware is working as hard as you are. Give your creative workflows a serious speed boost today and start creating your AI art more efficiently.

🆔 @hackernewsgithubprojects

📦 kuku-mom/kuku

Kuku: The Ultimate Local-First Markdown Workspace

Stop locking your notes away in closed services and reclaim your digital workspace with Kuku. This powerful, open-source Markdown app keeps your files local, private, and fully portable. Whether you are building a personal wiki, organizing your Second Brain, or using AI to draft and refine your thoughts, Kuku ensures your data remains yours alone. It features seamless encrypted syncing, 2D and 3D graph visualizations, and AI-native editing that lets you review changes before they are applied. Experience a knowledge workspace that puts you back in control—download Kuku and start building your own self-hosted, private intelligence hub today.

🆔 @hackernewsgithubprojects

📦 laksh-ya/offpayapp

Make Offline UPI Payments with OffPay

Tired of needing the internet just to send money? OffPay is an Android application that lets you perform UPI payments, check your bank balance, and scan QR codes entirely offline by leveraging your phone’s existing voice channel. Instead of struggling with complex raw dialer codes, the app provides a polished, modern interface that automates the experience for you. With features like encrypted transaction history, zero tracking, and a focus on privacy, OffPay ensures your data remains on your device. Whether you need an auto-filling assistant or a simple manual mode, it is a clever way to bank anywhere.

🆔 @hackernewsgithubprojects

📦 saard00/ai-youtube-shorts-generator

Automate Your Faceless YouTube Shorts with AI

Tired of spending hours editing videos for your channel? This automated Python pipeline builds engaging faceless Shorts entirely from a single topic. It uses Google Gemini to craft compelling scripts and Suno Bark for expressive narration. The system even creates a unique dual-visual style by pulling stock footage from Pexels and applying professional edits like scene splits and dynamic transitions with FFmpeg. It can even inject a custom mascot avatar to build your brand identity automatically. This is a complete production factory that handles everything from research to the final export, making high-quality content creation faster than ever before.

🆔 @hackernewsgithubprojects