Github Top Repositories

⚡ moeru-ai/airi is making waves. Here's the full picture. 🔗 https://github.com/moeru-ai/airi 📝 💖🧸 Self hosted, you-owned Grok Companion, a container of souls of waifu, cyber livings to bring them into our worlds, wishing to achieve Neuro-sama's altitude. Capable of realtime voice chat, Minecraft, Factorio playing. Web / macOS / Windows supported. ────────────────────────────── Project AIRI is an open-source project that aims to re-create Neuro-sama, a virtual character that can interact with users in a more human-like way. The project's goal is to provide a digital companion that can play games, chat, and perform various tasks, making it a unique and engaging experience. Key features of Project AIRI include its use of modern web technologies such as WebGPU, WebAudio, and WebAssembly, allowing for a more immersive and interactive experience. The project also has a strong focus on community involvement, with a dedicated Discord server and a translation project on Crowdin. To get started with Project AIRI, users can download the latest release for their operating system or try it out in their web browser. The project also provides a list of devlogs that detail the latest updates and changes. From a technical standpoint, Project AIRI is built using a range of technologies, including Web Workers, WebSocket, and NVIDIA CUDA. The project's use of these technologies allows for a high level of performance and customization. The project is suitable for a wide range of users, including developers, gamers, and anyone interested in AI and virtual companions. Takeaway: With Project AIRI, you can have your own digital companion that can play games, chat, and interact with you in a more human-like way, anywhere and anytime. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🔥 Leonxlnx/taste-skill is trending — and it deserves your attention. 🔗 https://github.com/Leonxlnx/taste-skill 📝 Taste-Skill - gives your AI good taste. stops the AI from generating boring, generic slop ────────────────────────────── The Taste Skill repository on GitHub offers a unique set of tools for developers looking to enhance their frontend designs with the help of AI agents. At its core, Taste Skill is designed to provide agent skills that improve layout, typography, motion, and spacing in AI-built interfaces, moving away from boilerplate-looking UIs. The repository includes both implementation skills that output code and image-generation skills for creating reference images. These skills can be easily installed using the npx skills add command, allowing developers to integrate them into their projects seamlessly. One of the key features of Taste Skill is its flexibility and customization options. Developers can choose from a variety of skills, each with its own specific focus, such as design-taste-frontend for a general, all-around frontend design or gpt-taste for stricter, GPT/Codex-oriented design rules. Additionally, skills like image-to-code-skill enable an image-first workflow where images are generated, analyzed, and then implemented in code. The technical highlights of Taste Skill include its framework-agnostic approach, making it compatible with major coding frameworks like React, Vue, and Svelte. The skills are also designed to be portable, using SKILL.md files that can be easily loaded by agents or copied into projects and conversations. For those interested in image generation, skills like imagegen-frontend-web and imagegen-frontend-mobile produce design images that can be used with ChatGPT Images, Codex, or other agents that generate images. Taste Skill is aimed at developers and designers looking to elevate their frontend designs with the assistance of AI. It's particularly useful for those working on premium frontend projects where layout, typography, and motion are crucial. By providing a set of specialized and adjustable skills, Taste Skill helps in creating unique, high-quality designs that stand out from the typical AI-generated UIs. To get started, developers can install the skills using npx skills add https://github.com/Leonxlnx/taste-skill, and then explore the various skills and their applications. For feedback and contributions, the community is invited to open issues or pull requests on GitHub or reach out directly to the maintainers. In summary, Taste Skill is a powerful tool for anyone looking to enhance their frontend design capabilities with AI, offering a range of skills and customization options to fit different project needs. With its ease of use, flexibility, and focus on high-quality design, Taste Skill is set to make a significant impact in the world of frontend development. So, take your designs to the next level with Taste Skill - where AI meets design excellence. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🔥 anthropics/claude-cookbooks is trending — and it deserves your attention. 🔗 https://github.com/anthropics/claude-cookbooks 📝 A collection of notebooks/recipes showcasing some fun and effective ways of using Claude. ────────────────────────────── The Claude Cookbooks GitHub repository provides a collection of code snippets, guides, and examples to help developers build with Claude, a powerful AI assistant. The repository offers a wide range of recipes for various capabilities, including classification, retrieval augmented generation, and summarization. It also covers tool use and integration, third-party integrations, multimodal capabilities, and advanced techniques. To get started, you'll need a Claude API key, which can be obtained for free. The code examples are primarily written in Python, but the concepts can be adapted to other programming languages. The repository is organized into sections, including: - Capabilities: Explore techniques for text and data classification, retrieval augmented generation, and summarization. - Tool Use and Integration: Learn how to integrate Claude with external tools and functions. - Third-Party Integrations: Supplement Claude's knowledge with external data sources. - Multimodal Capabilities: Discover how to use Claude with vision and generate images. - Advanced Techniques: Learn about sub-agents, uploading PDFs, automated evaluations, and more. The Claude Cookbooks are designed for developers who want to build with Claude and create innovative AI-powered applications. Whether you're a seasoned developer or just starting out, this repository provides a valuable resource for exploring the capabilities of Claude. One-liner takeaway: Unlock the full potential of Claude with the Claude Cookbooks, your go-to resource for building innovative AI-powered applications. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🚀 Meet paperless-ngx/paperless-ngx: a gem from today's GitHub trending list. 🔗 https://github.com/paperless-ngx/paperless-ngx 📝 A community-supported supercharged document management system: scan, index and archive all your documents ────────────────────────────── Paperless-ngx is a document management system that helps you digitize your physical documents, making them searchable and easily accessible online. The system is the official successor to the original Paperless and Paperless-ng projects, designed to be community-driven and supported by a team of people. Key features of Paperless-ngx include a user-friendly interface, document scanning and uploading, tagging and categorization, and search functionality. The system is highly customizable, with a docker compose installation option for easy deployment. To get started with Paperless-ngx, you can use the docker compose installation method or follow the step-by-step guides in the documentation. The system is also available for migration from Paperless-ng, with a simple drop-in replacement process. From a technical standpoint, Paperless-ngx is built using a range of technologies, including Python and JavaScript. The system is designed to be scalable and secure, with a focus on community involvement and contribution. The target audience for Paperless-ngx includes individuals and organizations looking to digitize their documents and streamline their workflow. The system is particularly useful for those who need to manage large volumes of documents, such as businesses, law firms, and government agencies. Overall, Paperless-ngx is a powerful and flexible document management system that can help you transform your physical documents into a searchable online archive. With its user-friendly interface and customizable features, Paperless-ngx is an ideal solution for anyone looking to go paperless - Ditch the paper, not the memories! ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

⚡ Fincept-Corporation/FinceptTerminal is making waves. Here's the full picture. 🔗 https://github.com/Fincept-Corporation/FinceptTerminal 📝 FinceptTerminal is a modern finance application offering advanced market analytics, investment research, and economic data tools, designed for interactive exploration and data-driven decision-making in a user-friendly environment. ────────────────────────────── Fincept Terminal is a state-of-the-art financial intelligence platform that offers institutional-grade financial analytics, AI automation, and unlimited data connectivity. Its key features include multi-asset analytics, AI agents, real-time trading, and global intelligence. The platform is built using C++20 and Qt6 for UI and rendering, with embedded Python for analytics. To use Fincept Terminal, you can download the installer for your platform or build it from source using CMake and Ninja. The platform is suitable for financial professionals, quantitative analysts, and developers who need advanced financial analytics and automation capabilities. With its extensive feature set and high-performance capabilities, Fincept Terminal is an ideal choice for those who want to take their financial analysis to the next level. The Fincept Terminal is the ultimate tool for financial Insights. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🔥 multica-ai/andrej-karpathy-skills is trending — and it deserves your attention. 🔗 https://github.com/multica-ai/andrej-karpathy-skills 📝 A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls. ────────────────────────────── The Andrej Karpathy Skills GitHub repository provides guidelines for improving Claude Code behavior, inspired by Andrej Karpathy's observations on LLM coding pitfalls. The project addresses issues like wrong assumptions, overcomplication, and lack of clarity in code. Key features include four principles: 1. Think Before Coding - explicit reasoning and questioning assumptions, 2. Simplicity First - minimal code and no overengineering, 3. Surgical Changes - touching only necessary code, 4. Goal-Driven Execution - defining success criteria and looping until verified. These principles are designed to be used by developers working with LLMs, particularly those using Claude Code. The guidelines can be installed as a plugin or added to a project's CLAUDE.md file. Technical highlights include the use of success criteria to transform imperative tasks into verifiable goals and the emphasis on simplicity and caution over speed. To get started, users can install the guidelines as a Claude Code plugin or add them to their project's CLAUDE.md file. The target audience is developers and teams working with LLMs and Claude Code, looking to improve their coding workflow and reduce mistakes. In summary: Follow these guidelines to make your LLMs loop until they get it right - success criteria and verification are key to efficient coding with LLMs. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

💡 manaflow-ai/cmux just hit the trending charts — here's why it matters. 🔗 https://github.com/manaflow-ai/cmux 📝 Ghostty-based macOS terminal with vertical tabs and notifications for AI coding agents ────────────────────────────── cmux is a macOS terminal designed for AI coding agents, built on top of Ghostty and offering features like vertical tabs, notifications, and an in-app browser. Key features include: - Notification rings and a notification panel for managing agent notifications - An with a scriptable API - Vertical and horizontal tabs for organizing workspaces - SSH support for remote machine workspaces - Claude Code Teams integration - Custom commands and scriptable functionality via the CLI and socket API To get started, you can download the DMG or install via Homebrew. Audience: Developers working with AI coding agents, particularly those using Claude Code and other similar tools. Technical highlights: Native macOS app built with Swift and AppKit, GPU-accelerated using libghostty. Give a million developers composable primitives like cmux and they'll build the most efficient workflows - faster than any product team. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

📌 Spotted on GitHub Trending: colbymchenry/codegraph — let's break it down. 🔗 https://github.com/colbymchenry/codegraph 📝 Pre-indexed code knowledge graph for Claude Code, Codex, Cursor, OpenCode, and Hermes Agent — fewer tokens, fewer tool calls, 100% local ────────────────────────────── Introducing CodeGraph, a revolutionary tool that supercharges your development workflow with semantic code intelligence. It integrates seamlessly with popular agents like Claude Code, Cursor, Codex CLI, opencode, and Hermes Agent, providing a pre-indexed knowledge graph that enables instant code exploration and search. To get started, simply run the installer with npx @colbymchenry/codegraph or use the provided scripts for macOS, Linux, or Windows. Then, initialize your project with codegraph init -i to build the knowledge graph index. Key Features include smart context building, full-text search, impact analysis, and framework-aware routes. CodeGraph supports 19+ languages and is 100% local, with no data leaving your machine. The benefits are clear: 35% cheaper, 57% fewer tokens, 46% faster, and 71% fewer tool calls. With CodeGraph, you can supercharge your coding experience. Try it today and discover a whole new world of coding efficiency - CodeGraph: because coding just got a whole lot smarter! ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🚀 Meet mukul975/Anthropic-Cybersecurity-Skills: a gem from today's GitHub trending list. 🔗 https://github.com/mukul975/Anthropic-Cybersecurity-Skills 📝 754 structured cybersecurity skills for AI agents · Mapped to 5 frameworks: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND & NIST AI RMF · agentskills.io standard · Works with Claude Code, GitHub Copilot, Codex CLI, Cursor, Gemini CLI & 20+ platforms · 26 security domains · Apache 2.0 ────────────────────────────── The Anthropic Cybersecurity Skills repository provides a comprehensive library of 754 production-grade cybersecurity skills across 26 security domains. This open-source project aims to equip AI agents with the skills of a senior security analyst, enabling them to perform tasks such as threat hunting, incident response, and vulnerability management. The repository includes five framework mappings, including MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, MITRE D3FEND, and NIST AI RMF, making it a unique and valuable resource for the cybersecurity community. To get started, users can clone the repository or use the npx skills add mukul975/Anthropic-Cybersecurity-Skills command. The skills are designed to be used with various AI platforms, including Claude Code, GitHub Copilot, and OpenAI Codex CLI. The project is not affiliated with Anthropic PBC and is an independent, community-created initiative. The repository is licensed under Apache-2.0 and is open to contributions. Overall, the Anthropic Cybersecurity Skills repository has the potential to significantly enhance the capabilities of AI agents in the cybersecurity domain, and its open-source nature makes it an exciting development for the community. Give your AI agent the security skills of a senior analyst — and watch it become a game-changer in cybersecurity. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🚀 Meet affaan-m/ECC: a gem from today's GitHub trending list. 🔗 https://github.com/affaan-m/ECC 📝 The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond. ────────────────────────────── The ECC (Efficient Computing Companion) repository is a comprehensive system for agentic work, built on top of various AI agent harnesses like Claude Code, Codex, and OpenCode. With over 182K stars and 28K forks, it's a widely-recognized project that's been developed over 10 months of intensive daily use. ECC provides a range of features, including skills, instincts, memory optimization, continuous learning, and security scanning. The system is production-ready and works across multiple harnesses, making it a versatile tool for developers. The project is constantly evolving, with new features and updates being added regularly. The most recent version, v2.0.0-rc.1, includes a public surface refresh, operator workflows, and an ECC 2.0 alpha. The system has a large community of contributors and users, with over 170 contributors and 12 language ecosystems supported. Whether you're a developer, researcher, or simply interested in AI, ECC is definitely worth checking out. With its comprehensive guides and extensive documentation, you'll be able to get started quickly and easily. So why wait? Dive into the world of ECC today and discover the power of efficient computing companions. The ECC repository is a one-stop-shop for all your agentic work needs - it's the ultimate tool for anyone looking to streamline their workflow and boost productivity. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

💡 rohitg00/ai-engineering-from-scratch just hit the trending charts — here's why it matters. 🔗 https://github.com/rohitg00/ai-engineering-from-scratch 📝 Learn it. Build it. Ship it for others. ────────────────────────────── The GitHub repository rohitg00/ai-engineering-from-scratch is a comprehensive, open-source curriculum designed to teach AI engineering from scratch. The purpose of this repository is to provide a structured approach to learning AI, covering 20 phases and 435 lessons that span ~320 hours of content. Key features of this curriculum include its focus on building AI systems from raw math, using languages like Python, TypeScript, Rust, and Julia. Each lesson follows a consistent structure, starting with the problem, deriving the math, writing the code, running the test, and keeping the resulting artifact. The curriculum is designed for individuals who want to understand how AI actually works, not just call APIs. It's suitable for those who can write code in any language, with some familiarity with Python being helpful. To get started, users can choose from three options: reading completed lessons on the website, cloning and running the repository, or finding their level using a placement quiz. Every lesson ships with a reusable tool, such as prompts, skills, agents, or MCP servers, which can be installed or pasted into daily workflows. The takeaway: Master AI engineering by building it from scratch, and become a part of a community that's shaping the future of artificial intelligence. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe