Github Top Repositories

📌 Spotted on GitHub Trending: rohitg00/ai-engineering-from-scratch — let's break it down. 🔗 https://github.com/rohitg00/ai-engineering-from-scratch 📝 Learn it. Build it. Ship it for others. ────────────────────────────── The AI Engineering from Scratch curriculum is designed to bridge the gap between using AI tools and understanding how they work. With 435 lessons and 20 phases, this free, open-source resource covers everything from math foundations to deep learning and autonomous systems. Each lesson follows a consistent structure, where you build an algorithm from scratch, then use it through a production library, resulting in a reusable artifact that you can apply to real-world problems. The curriculum is suitable for anyone who wants to understand how AI works, not just call APIs. By the end, you'll have a portfolio of 435 artifacts that you actually understand because you built them. Build it, use it, ship it - that's the motto, and with this curriculum, you'll be well on your way to becoming an AI engineer from scratch. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🔥 ChromeDevTools/chrome-devtools-mcp is trending — and it deserves your attention. 🔗 https://github.com/ChromeDevTools/chrome-devtools-mcp 📝 Chrome DevTools for coding agents ────────────────────────────── Introducing ChromeDevTools/chrome-devtools-mcp, a game-changing tool that lets your coding agent control and inspect a live Chrome browser. This Model-Context-Protocol (MCP) server gives your AI coding assistant access to the full power of Chrome DevTools, enabling reliable automation, in-depth debugging, and performance analysis. The key features of chrome-devtools-mcp include: - Getting performance insights by recording traces and extracting actionable performance insights - Advanced browser debugging capabilities, such as analyzing network requests, taking screenshots, and checking browser console messages - Reliable automation using puppeteer to automate actions in Chrome and automatically wait for action results To use chrome-devtools-mcp, you'll need Node.js, Chrome, and npm. The tool supports a range of coding agents, including Antigravity, Claude, Cursor, and Copilot, and provides a CLI for use without MCP. From a technical standpoint, chrome-devtools-mcp is built using Node.js and leverages the power of Chrome DevTools and puppeteer. The tool is designed to be flexible and customizable, with options for slim mode and performance analysis. chrome-devtools-mcp is perfect for developers who want to supercharge their coding workflow with the power of Chrome DevTools. Whether you're looking to automate tasks, debug issues, or analyze performance, this tool has got you covered. In short, chrome-devtools-mcp is a must-have tool for any developer looking to take their coding skills to the next level - it's like having a superpower in your coding toolbox! ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

📌 Spotted on GitHub Trending: rmyndharis/OpenWA — let's break it down. 🔗 https://github.com/rmyndharis/OpenWA 📝 Free, Open Source, Self-Hosted WhatsApp API Gateway ────────────────────────────── Introducing OpenWA, a free, open-source WhatsApp API Gateway designed for developers who need full control over their messaging infrastructure. With a pluggable architecture, OpenWA allows you to swap database engines, storage backends, and cache layers without changing a single line of application code. The key features of OpenWA include a REST API, multi-session support, webhooks, a web dashboard, and API key authentication. It also supports text messages, media messages, message reactions, bulk messaging, and message status tracking. To get started with OpenWA, you can use the docker compose command to start the application in development mode or production mode. The project uses a Node.js runtime, NestJS framework, and TypeScript language. OpenWA is suitable for developers who want to build custom messaging applications without vendor lock-in or hidden paywalls. With its flexible architecture and comprehensive documentation, OpenWA is an ideal choice for businesses and individuals looking for a reliable and scalable WhatsApp API Gateway. Take control of your messaging infrastructure with OpenWA - the open-source WhatsApp API Gateway that puts you in the driver's seat! ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🚀 Meet HKUDS/CLI-Anything: a gem from today's GitHub trending list. 🔗 https://github.com/HKUDS/CLI-Anything 📝 "CLI-Anything: Making ALL Software Agent-Native" -- CLI-Hub:https://clianything.cc/ ────────────────────────────── CLI-Anything is a revolutionary project that bridges the gap between AI agents and the world's software by making any software agent-ready with just one command line. The project's purpose is to empower agents to interact with various software, enabling them to perform tasks autonomously. Key features of CLI-Anything include the CLI-Hub, a central registry where you can browse, search, and install any CLI with a single pip install cli-anything-hub command, followed by cli-hub install <name>. The project also provides demos that showcase AI agents using generated CLIs to produce real artifacts. To get started, users can visit the CLI-Hub to explore and install CLIs. The project is built using python, click, and pytest, with a focus on JSON and human-readable output. CLI-Anything is designed for developers, AI researchers, and power users who want to automate tasks and integrate various software with AI agents. With its extensive test coverage and security features, the project ensures reliable and secure interactions between agents and software. In summary, CLI-Anything is a game-changer for AI-agent interaction, and its potential is vast: one command line to rule them all. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🚀 Meet obra/superpowers: a gem from today's GitHub trending list. 🔗 https://github.com/obra/superpowers 📝 An agentic skills framework & software development methodology that works. ────────────────────────────── Superpowers is a software development methodology for coding agents, providing a set of composable skills and instructions to enhance their capabilities. The system starts by asking users about their goals, then breaks down the project into manageable chunks, and creates an implementation plan emphasizing test-driven development, simplicity, and code quality. Key features include test-driven-development, systematic-debugging, and collaboration tools, such as brainstorming, writing-plans, and executing-plans. To use Superpowers, simply install the plugin for your preferred coding agent, such as Claude Code, Codex CLI, or GitHub Copilot CLI. From a technical standpoint, Superpowers is built around a skills library that includes testing, debugging, and collaboration tools, all designed to work seamlessly across different coding agents. This project is suitable for developers and teams looking to streamline their development process and improve code quality. In short, Superpowers is an innovative approach to software development that can supercharge your coding agents and take your projects to the next level - automating the process, not just the code. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

⚡ dotnet/skills is making waves. Here's the full picture. 🔗 https://github.com/dotnet/skills 📝 Repository for skills to assist AI coding agents with .NET and C# ────────────────────────────── The dotnet/skills repository is a curated set of core skills and custom agents for coding agents, following the Agent Skills standard from agentskills.io. It includes plugins for various .NET tasks such as data access, performance investigations, build skills, and more. To use these skills, you can install them via the Copilot CLI, VS Code, or Cursor, by adding the marketplace and installing individual plugins. You can also install skills using the Codex CLI. The repository is perfect for .NET developers looking to automate tasks and improve their workflow. It's also a great resource for contributors who want to add new plugins or improve existing ones. The technical highlights of this repository include its compatibility with OpenAI Codex and its use of the agentskills.io standard. The takeaway: Unlock your .NET development potential with dotnet/skills and take your coding to the next level! ────────────────────────────── 🧠 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 multica-ai/andrej-karpathy-skills GitHub repository is inspired by Andrej Karpathy's observations on the pitfalls of Large Language Models (LLMs) in coding. The repo provides a CLAUDE.md file that outlines four key principles to improve Claude Code behavior: Think Before Coding, Simplicity First, Surgical Changes, and Goal-Driven Execution. These principles address common issues such as wrong assumptions, overcomplication, and unnecessary changes. The repo offers two installation options: as a Claude Code plugin or by adding the guidelines to a project's CLAUDE.md file. The guidelines are designed to be merged with project-specific instructions and can be customized to fit individual needs. The ultimate goal is to reduce costly mistakes and promote efficient coding practices. One key insight from Andrej Karpathy is that LLMs excel at looping until they meet specific goals, so giving them success criteria and watching them go is a powerful approach. In a nutshell, this repo helps you code smarter, not harder, with LLMs - so, give your LLMs a compass, not a roadmap. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

💡 colbymchenry/codegraph just hit the trending charts — here's why it matters. 🔗 https://github.com/colbymchenry/codegraph 📝 Pre-indexed code knowledge graph for Claude Code, Codex, Cursor, and OpenCode — fewer tokens, fewer tool calls, 100% local ────────────────────────────── CodeGraph is a semantic code intelligence tool that supercharges Claude Code, Cursor, Codex, and OpenCode by providing a pre-indexed knowledge graph of codebases. This graph contains symbol relationships, call graphs, and code structure, allowing agents to query it instantly instead of scanning files. Key features include smart context building, full-text search, impact analysis, and always-fresh file watching. It supports 19+ languages and is framework-aware, recognizing web-framework routing files and emitting route nodes linked to their handlers. To get started, run the interactive installer with npx @colbymchenry/codegraph, then restart your agent and initialize projects with codegraph init -i. Technical highlights include a 35% cheaper and 70% fewer tool call rate, with a 100% local database and no external services required. The target audience includes developers using Claude Code, Cursor, Codex, or OpenCode, who want to improve their code exploration experience. In short, CodeGraph is a game-changer for code exploration - it's like having a superpower for your code editor, making you code smarter, not harder. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🚀 Meet anthropics/claude-plugins-official: a gem from today's GitHub trending list. 🔗 https://github.com/anthropics/claude-plugins-official 📝 Official, Anthropic-managed directory of high quality Claude Code Plugins. ────────────────────────────── The anthropics/claude-plugins-official GitHub repository is a curated directory of high-quality plugins for Claude Code. It features internal plugins developed by Anthropic and third-party plugins from partners and the community. To install a plugin, users can run /plugin install {plugin-name}@claude-plugins-official or browse the plugin marketplace in /plugin > Discover. From a technical standpoint, each plugin follows a standard structure, with required plugin.json metadata and optional components like commands/, agents/, and skills/. The repository is designed for developers and users looking to extend Claude Code's functionality. Overall, this repository is a one-stop shop for Claude Code plugins - install, explore, and create to unlock the full potential of Claude Code. You can't have too many tools in your toolbox, especially when it comes to coding. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🌟 HKUDS/ViMax caught my eye on GitHub Trending today. 🔗 https://github.com/HKUDS/ViMax 📝 "ViMax: Agentic Video Generation (Director, Screenwriter, Producer, and Video Generator All-in-One)" ────────────────────────────── Hi there, junior dev. Let's dive into the ViMax GitHub repo. ViMax is an innovative video generation tool that tackles current limitations in AI-generated videos, such as limited clip lengths, inconsistent characters and scenes, and a lack of narrative depth. The ViMax solution is an all-in-one platform that handles scriptwriting, storyboarding, character creation, and final video generation. You simply input your concept, and ViMax takes care of the rest. Some key features of ViMax include: - Idea2Video: transforms raw ideas into complete video stories - Novel2Video: adapts complete novels into episodic video content - Script2Video: creates videos from any screenplay - AutoCameo: generates videos from your photos From a technical standpoint, ViMax boasts an intelligent long script generation engine, expressive storyboard design, and automated quality control. The system is designed to eliminate production bottlenecks, providing effortless production, complete creative freedom, and professional-quality output. ViMax is perfect for creators who want to bring their ideas to life without worrying about the technical complexities of video production. So, what are you waiting for? Join the ViMax community and start creating your own videos today - with ViMax, your imagination is the only limit. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🎯 ggml-org/llama.cpp landed on trending. Worth a proper look. 🔗 https://github.com/ggml-org/llama.cpp 📝 LLM inference in C/C++ ────────────────────────────── The ggml-org/llama.cpp GitHub repository is a C/C++ implementation of LLM inference, focused on delivering high-performance and minimal setup. The project supports various models, including LLaMA, Mistral, and more. Key features include plain C/C++ implementation, Apple silicon optimization, and support for multiple architectures like x86, RISC-V, and GPUs. Technical highlights include 1.5-bit to 8-bit integer quantization, custom CUDA kernels, and Vulkan and SYCL backend support. The project also offers CPU+GPU hybrid inference for models larger than the total VRAM capacity. llama.cpp has a wide range of supported models, from text-only to multimodal models like LLaVA and Obsidian. To get started, you can install llama.cpp using brew, nix, or winget, or run it with Docker. Once installed, you can use the llama-cli command to work with local models or download models directly from Hugging Face. The project also offers a REST API server and various bindings for languages like Python, Go, and Node.js. The target audience includes developers, researchers, and anyone interested in LLM inference. Whether you're looking to integrate LLMs into your project or simply want to explore the capabilities of llama.cpp, this repository provides a comprehensive and well-maintained solution. With its high-performance capabilities, minimal setup, and wide range of supported models, llama.cpp is an excellent choice for anyone looking to dive into LLM inference. So, get started with llama.cpp today and discover the power of LLMs for yourself - llama.cpp is the fast lane to LLM inference! ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🌟 rohitg00/agentmemory caught my eye on GitHub Trending today. 🔗 https://github.com/rohitg00/agentmemory 📝 #1 Persistent memory for AI coding agents based on real-world benchmarks ────────────────────────────── The agentmemory GitHub repository provides a solution for coding agents to retain persistent memory, eliminating the need for re-explaining previous interactions. This tool is built on the iii engine and supports various agents, including Claude Code, Codex CLI, and Gemini CLI. agentmemory silently captures an agent's actions, compresses them into searchable memory, and injects relevant context into subsequent sessions. To use agentmemory, simply install it via npm install -g @agentmemory/agentmemory and start the memory server with agentmemory. The tool works with any agent that supports hooks, MCP, or REST API, and shares memories across all connected agents. The repository highlights several key features, including 95.2% retrieval accuracy, 92% fewer tokens, and support for 53 MCP tools. It also includes a detailed table showcasing the agents it works with, such as Claude Code, Codex CLI, and OpenClaw. In terms of technical highlights, agentmemory has 0 external databases, 12 auto hooks, and over 950 passing tests. The repository provides a demo.gif to illustrate its functionality and includes a benchmarking section that compares its retrieval accuracy to a grep baseline. The target audience for this tool appears to be developers working with coding agents, particularly those using the iii engine. With agentmemory, they can streamline their workflow and reduce the need for repetitive explanations. In summary, agentmemory is a game-changer for coding agents, enabling them to learn from previous interactions and providing a more efficient workflow - say goodbye to re-explaining everything, every time. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

📌 Spotted on GitHub Trending: zakirullin/files.md — let's break it down. 🔗 https://github.com/zakirullin/files.md 📝 🌱 Private, quiet space for thinking. A simple app for your .md files. ────────────────────────────── Files.md is a simple, private, and local-first app for storing your notes, documents, journal, habits, checklists, and tasks in plain .md files. This app is designed to be a quiet space for thinking, with a focus on local-first and privacy, meaning your data is never sent to a server. You can use Files.md to store your whole life, including notes, documents, journal records, and checklists. The app is free and open-source, with extremely simple code that can be understood and extended by a single person or an LLM. Key features include the ability to work offline, optional out-of-the-box synchronization, and a Telegram chatbot for on-the-go access to your files. To get started, simply open app.files.md in your browser, click "Install files.md," and open a local folder to persist changes. Files.md is perfect for individuals who want to take control of their data and think deeply about their notes and ideas. By using this app, you can connect ideas, let them compound, and think through your notes to gain a deeper understanding of the world. The app is suitable for anyone looking for a simple, private, and flexible note-taking solution, including writers, researchers, students, and individuals looking to improve their productivity and thinking skills. In short, Files.md is a powerful tool for anyone looking to take control of their data and think deeply about their ideas - own your data, own your thoughts. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

⚡ opentoonz/opentoonz is making waves. Here's the full picture. 🔗 https://github.com/opentoonz/opentoonz 📝 OpenToonz - An open-source full-featured 2D animation creation software ────────────────────────────── OpenToonz is a 2D animation software based on Toonz Studio Ghibli Version, which was originally developed by Digital Video, Inc. and customized by Studio Ghibli. Key features include a comprehensive set of tools for creating and editing 2D animations. To get started, you can download and install OpenToonz from the latest installer on the OpenToonz site. From a technical perspective, OpenToonz is built using a variety of technologies and is released under the Modified BSD License, allowing for free use and modification for business or personal purposes. The software has a strong community behind it, with a Google forum for users and a GitHub issues page for developers and bug reporters. The target audience for OpenToonz includes professional animators, students, and hobbyists looking for a powerful and customizable 2D animation tool. Overall, OpenToonz is a powerful and flexible 2D animation software that's free to use and modify. OpenToonz is a game-changer for animators - and it's completely free. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe