📦 swarmclawai/swarmvault

Build Your Own AI Second Brain with SwarmVault

Stop drowning in scattered notes and files! SwarmVault is a local-first LLM Wiki that builds an intelligent knowledge graph from your documents, code, and transcripts. It acts as an AI-powered second brain, transforming raw data into a structured, searchable knowledge base without needing constant API keys. Whether you are managing research, personal projects, or code repositories, it helps you visualize connections between ideas effortlessly. With its robust conflict detection and easy command-line interface, you can finally maintain a durable, evergreen memory store. Get started today with one simple command and turn your chaotic information into a powerful, automated research assistant.

🆔 @hackernewsgithubprojects

📦 suanmosuanyangtechnology/memorybear

MemoryBear: Giving AI a Human-Like Memory

Stop treating AI memory like a static filing cabinet and start giving it a brain that actually evolves. MemoryBear is a groundbreaking project that equips artificial intelligence with a human-like memory lifecycle including perception, extraction, association, and forgetting. By using a graph-first architecture with Neo4j and a sophisticated self-reflection engine, it mimics biological brain functions to prune redundant data and strengthen high-value knowledge. Whether you need to solve context window limitations or bridge data silos between agents, this system transforms passive AI into a proactive cognitive partner. Experience the next generation of intelligent memory management today.

🆔 @hackernewsgithubprojects

📦 vostride/agent-qa

Stop QA Flakiness with Agentic Memory

Are you tired of flaky tests that break every time your UI changes? Meet agent-qa, a self-improving quality assurance harness that brings memory to your testing workflow. Instead of rigid scripts, you write tests in natural language for both web and mobile applications. When a test fails, the system automatically observes the screen and attempts a different path, effectively healing itself during execution. It even remembers past runs to prevent recurring errors. By combining a polished dashboard with automated test authoring, it helps you catch regressions before they ship. Simplify your testing and let your agents handle the heavy lifting.

📰 https://news.ycombinator.com/item?id=48191312

🆔 @hackernewsgithubprojects

📦 kohya-ss/comfyui-anima-lllite

Master Anima ControlNet with LLLite in ComfyUI

Unlock powerful control over your Anima DiT models directly inside ComfyUI. This repository provides a minimal, efficient reference implementation for LLLite, a lightweight ControlNet variant that injects low-rank corrections into your model. It simplifies your workflow by automatically reading architectural settings from your weight files, supporting both standard and 4-channel inpainting tasks. You can even chain multiple nodes together for complex conditions without any conflict, thanks to its smart cascading wrapper system. If you need a flexible, streamlined way to apply precise control to your generative projects, this tool is the perfect, ready-to-use addition to your library.

🆔 @hackernewsgithubprojects

📦 zhengdian1/interleavethinker

Supercharge Your AI Image Generation with InterleaveThinker

Ever wish your AI image generator could plan and critique its own creative process? Meet InterleaveThinker, a cutting-edge multi-agent pipeline that adds sophisticated reasoning to any existing image generation model. By using a specialized planner to organize sequences and a critic agent to identify errors, it enables complex, multi-step visual narratives and task annotations. It even includes training datasets and a reinforcement learning framework to refine instruction accuracy, helping you achieve impressive, high-quality results for challenging visual tasks. If you want to push the boundaries of what your AI can create, give your models the power of thoughtful iteration today.

🆔 @hackernewsgithubprojects

📦 samvallad33/vestige

Give Your AI a Real Memory with Vestige

Does your AI forget project details the moment you start a new session? Meet Vestige, a local cognitive memory system designed to help your AI agents remember what actually matters. Instead of basic data storage, it uses neuroscience-inspired techniques like FSRS-6 spaced repetition, spreading activation, and memory dreaming to manage information just like a human brain. It runs as a single, efficient Rust binary and integrates directly with tools like Claude, Cursor, and Xcode. You can even visualize your AI's evolving knowledge through a real-time 3D dashboard. Stop dealing with amnesiac agents and start building smarter, persistent workflows today.

🆔 @hackernewsgithubprojects

📦 dimartarmizi/omnicloud

Manage All Your Cloud Storage In One Place With OmniCloud

Tired of jumping between multiple cloud storage accounts just to find your files? OmniCloud solves this by acting as a powerful full-stack aggregation platform that brings all your storage providers into one consistent workspace. Whether you use Google Drive, OneDrive, Dropbox, or S3-compatible storage, this tool lets you browse, upload, and manage files seamlessly from a single dashboard. With features like drag-and-drop uploads, real-time progress tracking, and smart storage allocation strategies, it simplifies your digital life significantly. Stop juggling browser tabs and take control of your data by centralizing your cloud ecosystem today with this efficient open-source project.

🆔 @hackernewsgithubprojects

📦 merserk/comfyui-pid

Upgrade Your ComfyUI Workflow with NVIDIA PiD

Stop settling for blurry AI outputs and start decoding your images with professional precision using NVIDIA PiD. This custom ComfyUI node brings the power of Pixel Diffusion Decoders directly into your creative pipeline, allowing you to turn low-resolution latents into sharp, detailed four-K images. By replacing traditional VAE decoding with a fast, four-step generative process, it natively handles complex upscaling without needing separate, heavy models. Whether you are working with Flux, SD3, or other compatible backbones, this toolkit provides a staged workflow to prepare, sample, and finalize your art with ease. Unlock superior image quality and transform your generation speed today.

🆔 @hackernewsgithubprojects

📦 ah4ddd/ascii-vision

Turn Any Image into Stunning ASCII Art!

Ever wanted to turn your photos into cool text-based art? This project is an advanced image-to-ASCII engine that transforms images into vibrant character-based visuals right in your browser. You can upload any photo and instantly apply different rendering modes like neon, scanline, or classic symbols. It even supports multiple languages, allowing you to use diverse Unicode glyph sets beyond standard keyboard characters. With built-in tools for adjusting contrast, brightness, and color palettes, you can fine-tune every detail before exporting your masterpiece as a text file or a high-quality image. Try it out to create your own unique digital art.

🆔 @hackernewsgithubprojects

📦 bryanyzhu/agentic-ai-system-course

Master AI Agents: The Ultimate Developer's Skeleton Course

Want to move beyond simple prompts and start building production-grade AI agents? This unique course provides the essential architecture you need to design autonomous systems that plan, use tools, and maintain memory. Instead of forcing you into a single framework, it functions as a flexible skeleton that you build upon with your own AI partner. Whether you are creating a coding assistant or a research tool, you will master patterns like multi-agent delegation, human-in-the-loop workflows, and robust memory management. Stop building in the dark and start engineering agents that actually work by learning to ask the right questions.

🆔 @hackernewsgithubprojects

📦 jqtangust/robust-u1

Can AI Fix Blurry Photos? Meet Robust-U1!

Ever wonder if AI could actually fix corrupted images instead of just guessing what they might be? Meet Robust-U1, an innovative model featured at ICML 2026 that changes the game for visual understanding. While traditional tools struggle with image degradation, this unified model goes a step further by self-recovering pixel-level details before performing its analysis. By utilizing a specialized three-stage training pipeline involving supervised fine-tuning and reinforcement learning, it bridges the gap between broken inputs and reliable insights. Dive into this project to see how it masters the art of visual self-recovery and robust reasoning for yourself today.

🆔 @hackernewsgithubprojects

📦 pocketpaw/pocketpaw

Build Your Own Private AI Agent Today

Are you tired of AI tools that compromise your privacy or lock you into their ecosystem? This project offers a powerful solution by letting you host your own personal AI agent right on your machine. With a simple desktop installer, it creates a secure command center where you can integrate models from providers like OpenAI, Anthropic, or even run local instances through Ollama. It features a multi-agent orchestration system and rigorous seven-layer security, ensuring your data remains yours alone. Take full control of your automation tasks and reclaim your productivity by building a truly private, self-hosted AI workspace today.

🆔 @hackernewsgithubprojects

📦 dl1683/irys-stateful-swarms

AI Agents That Remember: The Stateful Swarm Revolution

Most AI systems are trapped in a cycle of forgetting, wasting money by relearning the same information for every single task. Irys-stateful-swarms breaks this cycle by replacing disposable, stateless reasoning with persistent, structured analytical state. Instead of starting from scratch, it coordinates multiple agents using a shared blackboard that preserves every observation and calculation with full source attribution. This architecture turns affordable models into top-tier performers, hitting over seventeen percent on complex legal benchmarks at a fraction of the usual cost. It proves that superior performance comes from a smarter system architecture rather than just using more expensive models.

🆔 @hackernewsgithubprojects

📦 mr-beaver/tokencost

Stop Overpaying for AI: Track and Cut Your Costs Instantly

Are you accidentally burning money on AI tools like Claude Code or VS Code? You might be spending way more than you think, but now you can take control. This repository provides a powerful local proxy that acts as a middleman for your API requests, giving you a crystal-clear dashboard of your exact spending. It does more than just track; it automatically optimizes your requests using smart routing and prompt caching to slash your bills by up to eighty percent. It works entirely on your machine, ensuring your private prompts never leave your local environment. Start saving your budget today.

🆔 @hackernewsgithubprojects

📦 recursive-org/first-steps-toward-automated-ai-research

Automated AI Research: Cracking the Code

Imagine an AI that researches itself to push hardware to its absolute limit. This repository showcases artifacts from an automated AI research system that discovers highly optimized code and training strategies. You will find impressive results like a record-breaking speedrun for training a language model in just over seventy-seven seconds, alongside specialized GPU kernel implementations that outperform standard benchmarks. By sharing these experimental outcomes and training scripts, this project provides a fascinating look into how machines can autonomously improve their own efficiency. Explore these findings to see the future of automated performance optimization for yourself.

🆔 @hackernewsgithubprojects

📦 albertofettucini/council

Stop Trusting Just One AI: Meet Council

Are you tired of relying on a single AI model and wondering what you might be missing? Council is a native macOS application that changes the game by gathering a panel of different AI minds to answer your questions in parallel. Instead of juggling multiple chat windows, you get side-by-side responses that critique each other blindly, removing brand bias to reveal where they agree or diverge. With support for twelve different backends and a focus on privacy using your own keys, Council helps you weigh multiple perspectives before making your own informed decision. Take control of your AI workflow today.

🆔 @hackernewsgithubprojects

📦 texasbedouin/vibe-check

Stop Guessing, Start Building: The Ultimate AI Coding Guide for Beginners

Have a great app idea but no idea how to actually build it? Meet a powerful AI tool designed to be your patient, expert mentor. Instead of staring at a blank screen, this guide takes you from a vague spark of inspiration to a professional, buildable blueprint. It helps you map out user experiences, select the right technology, and break your project into manageable phases. It even includes a special mode to help clean up messy code as your app grows. Stop feeling overwhelmed and finally turn your vision into a real, functional project with confidence and clarity today.

🆔 @hackernewsgithubprojects

📦 nvlabs/spatialclaw

SpatialClaw: A Smarter Way for AI to Reason in 3D

Ever wonder how AI can better understand 3D spaces? SpatialClaw is a clever, training-free framework that fundamentally changes how vision-language models tackle spatial reasoning. Instead of being limited by rigid, pre-set commands, it treats code as the primary action interface. By using a persistent Python kernel equipped with powerful perception tools like segmentation and depth reconstruction, the agent can write and execute its own code step-by-step. This allows it to flexibly analyze visual evidence and adapt its strategy on the fly. It is a major leap in performance, proving that giving AI the power to code makes it much smarter.

🆔 @hackernewsgithubprojects

📦 0xovo/litedoc

Convert PDFs to Markdown Instantly—100% Locally!

Are you tired of paying for AI tools or uploading sensitive documents to external servers just to extract text? Meet LiteDoc, a completely private, browser-based solution that converts your PDFs into clean, GitHub-Flavored Markdown. Because it runs entirely on your local machine using your browser's resources, your files never leave your device. It handles complex layouts, scientific math formulas, and even scanned documents through local WebAssembly-powered OCR. Whether you are a student or a researcher, it is the perfect zero-setup tool to get your content ready for any AI pipeline. Take full control of your documents starting today.

🆔 @hackernewsgithubprojects

📦 ccpt5/comfyui-berninistudio

Simplify Your Bernini Video Edits in ComfyUI

Are you tired of messy ComfyUI workflows just to run Bernini video editing? This node is the ultimate shortcut, wrapping complex text encoding and latent construction into one single, clean node instead of wiring five different connections. It streamlines your process while adding powerful built-in Ollama and vLLM prompt enhancement. By leveraging official templates and even vision-capable models to analyze your reference images, it makes prompting more accurate than ever. It is a fantastic way to simplify your setup and level up your AI video projects. Give it a try to experience a much smoother, faster, and more efficient creative workflow.

🆔 @hackernewsgithubprojects