Claw Code is a high-performance Rust implementation of a CLI agent harness that integrates with AI models like Claude and OpenAI-compatible providers to enable automated code generation and task execution. Built by UltraWorkers and achieving viral adoption with nearly 200K stars, it provides developers with a powerful command-line tool for running AI-driven workflows, executing code, managing sessions, and integrating with local or cloud-based language models. The repository includes comprehensive documentation, Docker support, RAG (retrieval-augmented generation) capabilities, and cross-platform support for Windows, macOS, and Linux.

OpenClaw is a personal AI assistant platform that runs on your own devices and integrates with multiple messaging channels like WhatsApp, Telegram, Slack, Discord, and many others. Built with TypeScript, it provides a self-hosted alternative to cloud-based AI assistants with support for voice/audio on macOS/iOS/Android and customizable skills. The project emphasizes data ownership, local execution, and seamless integration across messaging platforms while maintaining security and extensibility through a plugin system.

Open Design is an open-source, local-first alternative to Claude Design that enables AI-powered design artifact creation using coding agents already installed on your machine. It integrates 16+ coding agent CLIs (Claude Code, Codex, Cursor, etc.) with 132+ composable skills and 150+ design systems to generate interactive web, desktop, mobile prototypes, slides, and other design artifacts through a skill-driven workflow. The platform runs as a native desktop app or web deployment, supports multiple export formats (HTML, PDF, PPTX, MP4), and maintains full BYOK (bring your own keys) capability at every layer without vendor lock-in.

ECC is a comprehensive agent operating system designed to optimize AI agent performance across multiple platforms including Claude Code, Cursor, Codex, OpenCode, and others. It provides a production-ready framework with 63+ agents, 249+ skills, hooks, rules, MCP configurations, and security features built from real-world multi-harness engineering workflows. The system includes token optimization, memory persistence, continuous learning capabilities, security scanning, and cross-harness agent orchestration to enable developers to build and manage sophisticated AI-powered applications.

Awesome DESIGN.md is a curated collection of 73+ DESIGN.md files extracted from popular brand websites and developer platforms, formatted as plain-text design system documents that AI agents can read to generate consistent UI. Each DESIGN.md file captures the design language, patterns, tokens, and visual rules of companies like Claude, Vercel, Linear, and Figma, enabling developers to drop them into projects and instruct AI agents to build matching interfaces. This resource is built on Google Stitch's DESIGN.md concept, making it easy for AI coding agents and design tools to understand how a project should look and feel without requiring Figma exports or complex configuration.

gstack is an open-source AI engineering toolkit that transforms Claude Code into a virtual team of 23 specialized roles (CEO, designer, engineer manager, QA lead, security officer, and more) through slash commands and Markdown-based workflows. Built by Garry Tan (President & CEO of Y Combinator), it enables solo developers and small teams to ship products at significantly accelerated rates by automating code review, testing, deployment, security audits, and architectural decisions. The toolkit includes a fast headless browser, comprehensive documentation, and integrations with multiple AI coding agents, making it accessible to founders, technical leaders, and first-time Claude Code users who want structured guidance for shipping at scale.

Awesome is the definitive curated list of awesome lists covering hundreds of topics across programming, platforms, learning resources, and more. Created and maintained by Sindre Sorhus, it serves as a centralized index for discovering high-quality, community-driven resource collections across virtually every domain of technology and beyond. This repository provides guidelines for creating and contributing to awesome lists, making it both a resource hub and a standard-setting template for the broader awesome list ecosystem.

freeCodeCamp is an open-source learning platform and curriculum that provides free, self-paced education in web development, machine learning, programming, and computer science to help adults transition into tech careers. Built with TypeScript, React, Node.js, and other modern technologies, the platform powers freeCodeCamp.org, offering interactive coding challenges, certifications, and a supportive community forum and Discord server. The repository contains the complete codebase for the learning platform, API, client interface, and curriculum content that has helped over 100,000 people land their first developer job.

Caveman is a Claude Code skill and AI agent plugin that dramatically reduces token usage by making AI assistants communicate in a terse, caveman-like style—cutting approximately 65-75% of output tokens while maintaining full technical accuracy. It works across 30+ AI coding agents (Claude Code, Codex, Gemini, Cursor, Copilot, etc.) and includes additional tools for compressing context files, generating commit messages, and tracking token savings. The project demonstrates that brief, fragmented responses can actually improve accuracy while making AI interactions faster and more cost-effective.

Public APIs is a community-curated comprehensive directory of free, public APIs across dozens of categories including weather, finance, health, entertainment, and more. Developers use this repository to discover and integrate free APIs into their projects without cost barriers. The repository is actively maintained by the open-source community and includes detailed information about each API's authentication requirements, HTTPS support, and CORS availability.

This repository maintains a comprehensive, community-curated list of free programming books, courses, podcasts, and screencasts available in multiple languages. Administered by the Free Ebook Foundation, it serves as an educational resource for learners and developers worldwide, with over 389,000 stars making it one of GitHub's most popular repositories. The project welcomes contributions from the community and provides easy-to-use search functionality and well-organized categorization by programming language, subject matter, and language translation.

Graphify is an AI coding assistant skill that transforms entire project folders—including code, databases, documentation, and multimedia—into interactive, queryable knowledge graphs. It integrates with popular AI coding assistants like Claude Code, Cursor, Codex, and Gemini CLI, enabling developers to understand complex codebases through graph visualization and natural language queries rather than traditional file searching. The tool generates an interactive HTML graph, markdown reports with key insights, and JSON data for further analysis, making it invaluable for code exploration, architecture understanding, and AI-assisted development workflows.

MemPalace is a local-first AI memory system that stores conversation history and project files as verbatim text with semantic search retrieval, achieving 96.6% recall on LongMemEval benchmarks without requiring any API calls or cloud services. Designed for developers and AI practitioners, it features a structured palace architecture (wings for people/projects, rooms for topics, drawers for content), pluggable backends like ChromaDB, MCP server integration for Claude and other tools, and knowledge graph capabilities—all running entirely on your machine with no data leaving unless explicitly opted in.

Roadmap.sh is a comprehensive community-driven platform that provides interactive, visual roadmaps for developers across 70+ career paths including frontend, backend, DevOps, data science, and specialized fields. The repository hosts the source code for an educational platform featuring interactive roadmaps, best practices guides, and knowledge assessment questions designed to help developers plan and progress their careers. Built with TypeScript and Astro, it serves millions of developers globally by offering structured learning paths and curated resources for various programming languages, frameworks, and specializations.

This repository contains a collection of reusable AI agent skills designed to improve software engineering practices when working with Claude and other coding agents. The skills address common failure modes like misalignment between developer and agent, verbose outputs, buggy code, and architectural degradation by implementing proven engineering practices such as clarifying requirements, establishing shared domain language, test-driven development, and systematic debugging. Created by Matt Pocock from his personal toolkit, these skills are modular, composable, and meant to be customized for any project, offering a practical alternative to rigid AI development frameworks that sacrifice control for convenience.

Superpowers is a comprehensive software development methodology and skills framework designed to enhance AI coding agents like Claude, GitHub Copilot, and Gemini. It provides a structured workflow that guides agents through brainstorming, planning, test-driven development, and systematic debugging—ensuring they work methodically rather than jumping directly to code. Available as plugins across multiple coding platforms, Superpowers enables agents to work autonomously for extended periods by composing reusable skills that enforce best practices like TDD, code review, and proper verification.

Career-Ops is an AI-powered job search automation system that leverages Claude Code to intelligently evaluate job opportunities, generate tailored ATS-optimized resumes, and manage the entire application pipeline. It features 14 skill modes, batch processing capabilities, a Go-based dashboard, PDF generation, and supports multiple AI providers (Claude, Gemini, OpenAI) across 45+ pre-configured job portals. Designed for job seekers who want to filter hundreds of opportunities down to the most promising few, the system uses structured evaluation criteria, interview prep story banking, and negotiation frameworks to turn job searching into a data-driven process.

Awesome Python is a curated, opinionated list of the best Python frameworks, libraries, tools, and resources organized across categories like AI/ML, web development, data science, and DevOps. It serves as a comprehensive guide for Python developers looking to discover high-quality tools and libraries, with an associated website (awesome-python.com) that enables easy searching and filtering. This repository is the #10 most-starred on GitHub and is maintained as a community-driven resource with sponsorship opportunities.

Awesome-Selfhosted is a comprehensive, community-maintained list of free and open-source software applications that can be self-hosted on personal servers instead of relying on SaaS providers. The repository covers hundreds of tools across dozens of categories including communication, file storage, content management, monitoring, and development tools, making it an essential resource for anyone interested in privacy-focused, independent hosting solutions. With nearly 300,000 stars, it serves both technical users and organizations seeking alternatives to commercial cloud services.

Paperclip is an open-source orchestration platform that helps teams manage and coordinate multiple AI agents to accomplish business goals. Built with TypeScript using a Node.js server and React UI, it provides features like org charts, budget management, governance controls, and task tracking—functioning as a central control plane for autonomous AI teams. Designed for users who work with multiple AI agents (Claude, Cursor, OpenClaw, etc.), Paperclip solves the coordination and cost management challenges that arise when running autonomous systems at scale.

React is a JavaScript library developed by Facebook for building user interfaces using a declarative, component-based approach. It enables developers to create interactive UIs efficiently by automatically updating and rendering components when data changes, with support for web applications, server-side rendering via Node.js, and mobile apps through React Native. The library is designed for gradual adoption and is MIT-licensed, making it one of the most popular and widely-used frontend frameworks in the industry.

The Linux kernel is the core component of Linux operating systems that manages hardware, system resources, and provides fundamental services for all software running on a system. This repository contains the complete source code for the Linux kernel and serves developers, researchers, system administrators, hardware vendors, and other contributors who want to build, modify, or maintain the kernel. It includes comprehensive documentation tailored to different user roles, from new kernel developers to security experts, along with tools for building, testing, and managing kernel contributions.

TheAlgorithms/Python is a comprehensive, community-driven repository that implements hundreds of algorithms in Python for educational purposes. It covers diverse domains including sorting, searching, dynamic programming, cryptography, machine learning, and more, serving as a learning resource for students, interview preparation, and algorithm practice. The project maintains high code quality standards with CI/CD pipelines, pre-commit hooks, and extensive documentation, making it an accessible reference for understanding algorithmic concepts and implementations.

Pretext is a pure JavaScript/TypeScript library for fast and accurate multiline text measurement and layout that avoids expensive DOM reflow operations. It implements its own text measurement logic using the browser's font engine and supports comprehensive multilingual text with features like proper line breaking, bidirectional text, and rendering to DOM, Canvas, and SVG. The library is designed for developers building web UIs who need precise text height calculations for virtualization, custom layouts, and preventing layout shifts without relying on traditional DOM measurements.

Understand Anything is an AI-powered plugin that transforms codebases, knowledge bases, and documentation into interactive, explorable knowledge graphs. It works across multiple AI coding platforms (Claude Code, Cursor, Copilot, Gemini CLI, and others) to help developers quickly grasp complex projects through visual navigation, semantic search, guided tours, and natural language queries. The tool generates architectural visualizations, domain maps, diff impact analysis, and onboarding guides—making it especially valuable for teams onboarding new members or navigating unfamiliar large codebases.

This repository contains Anthropic's implementation of Agent Skills for Claude, which are modular packages of instructions, scripts, and resources that dynamically enhance Claude's capabilities for specialized tasks. The repository provides example skills spanning creative design, technical development, enterprise workflows, and document manipulation, along with a specification for the Agent Skills standard and templates for creating custom skills. Developers can use these skills with Claude Code, Claude.ai, and the Claude API to teach Claude how to complete domain-specific tasks in a repeatable and consistent manner.

Hermes Agent is a self-improving AI agent built by Nous Research that features a built-in learning loop, creating and refining skills from experience while supporting any LLM provider (OpenAI, Claude, open-source models, etc.). It offers a feature-rich terminal interface, multi-platform messaging integration (Telegram, Discord, Slack, etc.), autonomous skill creation, scheduled automations, and can run anywhere from a $5 VPS to serverless infrastructure. Designed for both individual users and researchers, it provides batch trajectory generation, cross-session memory with full-text search, and parallel agent delegation without vendor lock-in.

This repository provides a comprehensive collection of algorithms and data structures implemented in JavaScript, complete with detailed explanations and links to further learning resources. It serves as an educational tool for computer science students, job interviewees, and developers seeking to understand and master fundamental algorithmic concepts. The repository covers topics ranging from basic data structures like linked lists and stacks to advanced algorithms including graph theory, dynamic programming, and cryptography, all thoroughly tested and documented.

TensorFlow is an open-source machine learning framework developed by Google that provides comprehensive tools and libraries for building and deploying ML-powered applications. It offers stable Python and C++ APIs, supports multiple platforms (Linux, Windows, macOS, Android, Raspberry Pi), and includes GPU acceleration capabilities for deep learning and neural network research. Designed for both researchers pushing state-of-the-art ML and developers building practical applications, TensorFlow is one of the most widely-used ML frameworks in the industry.

n8n is a fair-code workflow automation platform that combines visual no-code building with custom code capabilities, offering 400+ integrations and native AI support for technical teams. It enables users to create powerful automations while maintaining full control over their data through self-hosting or cloud deployment options. Built with TypeScript and Node.js, n8n serves as an accessible yet powerful alternative to traditional IPaaS and RPA tools for developers and non-technical users alike.

Oh My Zsh is a community-driven framework for managing zsh shell configuration, offering 300+ plugins and 140+ themes that enhance terminal productivity and aesthetics. Designed for developers and power users on macOS, Linux, Windows (WSL2), and other Unix-like systems, it provides easy installation, auto-updates, and extensive customization options to make the command-line experience more powerful and enjoyable. With nearly 200k stars and 2,500+ contributors, it has become one of the most popular shell configuration frameworks in the developer community.

Visual Studio Code is a lightweight yet powerful open-source code editor developed by Microsoft that combines simplicity with comprehensive development features including code editing, debugging, and extensibility. This repository hosts the Code - OSS source code under the MIT license, where Microsoft and the community collaboratively develop the product, publish roadmaps, and manage issues. It serves developers across Windows, macOS, and Linux who need a fast, feature-rich editor with built-in tools for the edit-build-debug cycle and extensive extension support.

AutoGPT is a comprehensive platform for building, deploying, and managing autonomous AI agents that automate complex workflows without requiring deep technical expertise. It provides both a low-code agent builder interface and a robust backend server, supporting multiple LLM providers (OpenAI, Claude, Llama), and offers pre-built agents ready for immediate use or customization through a visual workflow designer. The platform is available for self-hosting via Docker or as a cloud-hosted beta, making AI agent automation accessible to developers and non-technical users alike.

CLI-Anything is a framework that transforms any software application into an agent-native command-line interface, enabling AI agents like Claude, Cursor, and Pi to programmatically control desktop and web applications. It provides a comprehensive registry (CLI-Hub) of 18+ pre-built CLI harnesses for popular tools like Blender, AutoCAD, QGIS, Audacity, and more, with strong testing coverage (2,269+ passing tests) and a community-driven contribution model. Developers can use CLI-Anything to make their software controllable by AI agents, while AI developers gain standardized, structured access to powerful creative and productivity tools.

Flutter is Google's open-source SDK for building beautiful, fast, and natively compiled applications for mobile, web, and desktop platforms from a single codebase using the Dart programming language. It enables developers and organizations worldwide to create pixel-perfect user experiences with stateful hot reload, extensive widget libraries, and hardware-accelerated graphics powered by Skia and Impeller. Flutter supports iOS, Android, Windows, macOS, Linux, and web platforms while offering seamless integration with native code through FFI and platform channels.

Microsoft Activation Scripts (MAS) is an open-source activation tool for Windows and Office that provides multiple activation methods including HWID, Ohook, TSforge, and Online KMS. Designed for users seeking to activate Microsoft products, it offers both PowerShell and traditional batch script interfaces with built-in troubleshooting capabilities and comprehensive documentation. The project is actively maintained with a large community and provides support across multiple platforms including GitHub, Discord, and Reddit.

OpenClaude is an open-source CLI agent that brings coding workflows to any LLM provider, including OpenAI, Gemini, Ollama, GitHub Models, and 200+ compatible models. It unifies cloud and local AI backends under a single terminal interface with support for tool calling, MCP (Model Context Protocol), agents, streaming output, and vision capabilities. Developers can switch between providers seamlessly while maintaining consistent access to coding tools like bash, file operations, grep, and automated agents.

Bootstrap is the world's most popular front-end framework for building responsive, mobile-first web projects with HTML, CSS, and JavaScript. It provides a comprehensive toolkit of pre-built components, utilities, and styling that enables developers to rapidly develop modern, accessible websites and applications. The framework is available through multiple package managers (npm, yarn, Composer, NuGet) and includes extensive documentation, examples, and a large community.

This is GitHub's official collection of `.gitignore` templates for popular programming languages, frameworks, tools, and operating systems. The repository serves as the source for the `.gitignore` template chooser available on GitHub.com when creating new repositories, helping developers quickly set up proper file exclusion rules for their projects. It's organized into mainstream templates for common use cases, global templates for editors and tools, and community templates for specialized frameworks, making it a comprehensive resource for both beginners and experienced developers.

Awesome Go is a comprehensive, community-curated list of Go frameworks, libraries, and software tools organized by category, with over 173,000 stars on GitHub. It serves as a go-to resource for Go developers looking to discover high-quality packages and projects across various domains like web frameworks, databases, testing tools, and more. The repository includes automated quality checks, generates a static website for easy browsing, and welcomes community contributions to keep the list fresh and relevant.

Ollama is an open-source platform that makes it easy to run large language models locally on your machine, supporting popular models like Llama, Gemma, Qwen, DeepSeek, and many others across macOS, Windows, Linux, and Docker. It provides a simple CLI interface, REST API, and integrations with popular tools like VS Code, Claude, and web chat interfaces, enabling developers and users to build AI applications without relying on cloud services. Written in Go with support for GPU acceleration, Ollama democratizes access to powerful LLMs by making them accessible to anyone regardless of technical expertise.

OpenCode is an open-source AI coding agent that automates software development tasks through an intelligent assistant interface. It provides two main agents—a full-access "build" agent for development and a read-only "plan" agent for code analysis—accessible via CLI, desktop app, and web interface. The tool is designed for developers who want to leverage AI to accelerate coding workflows, with support for multiple platforms and extensive customization options.

yt-dlp is a feature-rich command-line tool for downloading audio and video from thousands of websites, including YouTube, with advanced features like SponsorBlock integration, flexible format selection, and subtitle support. It is a maintained fork of the inactive youtube-dl project, designed for users who need reliable, powerful media downloading capabilities across multiple platforms. The tool is actively developed with extensive customization options, plugin support, and cross-platform binary releases for Windows, macOS, and Linux.

Agent Skills is a collection of 23 production-grade engineering workflows and best practices packaged for AI coding agents like Claude, Cursor, and Gemini. It encodes senior engineer workflows across the entire development lifecycle—from specification and planning through testing, review, and deployment—with structured steps, quality gates, and verification procedures. The skills are accessible through slash commands and integrate with multiple AI agent platforms, enabling consistent application of software engineering best practices regardless of which AI tool developers use.

Transformers is a comprehensive Python framework for working with state-of-the-art pre-trained machine learning models across text, vision, audio, and multimodal domains. It serves as the unified model-definition standard that ensures compatibility across the broader ML ecosystem, enabling seamless integration with various training frameworks, inference engines, and modeling libraries. With over 1 million pre-trained model checkpoints available on the Hugging Face Hub, it democratizes access to cutting-edge AI models through simple, customizable, and efficient APIs. Designed for both researchers and practitioners, it supports both inference and training workflows across multiple deep learning frameworks.

RTK (Rust Token Killer) is a high-performance CLI proxy that dramatically reduces LLM token consumption by 60-90% on common development commands through intelligent filtering, compression, and output optimization. Built as a single zero-dependency Rust binary with sub-10ms overhead, it works seamlessly with AI coding agents like Claude Code, Cursor, and Windsurf by intercepting and reformatting command outputs before they reach the LLM context. The tool supports 100+ commands across version control, testing, linting, and file operations, making it invaluable for developers using AI coding assistants who want to minimize API costs while maintaining code understanding.

HelloGitHub is a curated monthly newsletter that discovers and shares interesting, beginner-friendly open source projects from GitHub, along with open source books, practical projects, and enterprise-level repositories. Published on the 28th of each month, it helps developers explore the open source ecosystem and find engaging projects to learn from through a multilingual platform (Chinese, English, and Japanese). The project is designed for developers of all levels who want to discover quality open source projects without spending excessive time searching.

JavaGuide is a comprehensive open-source interview preparation and learning resource for Java backend developers, covering Java fundamentals, computer science basics, databases, distributed systems, and high concurrency topics. With over 155k stars, it serves as a complete guide for job interview preparation and backend development skills, complemented by paid resources like AI application development guides and system design interview materials. The repository includes detailed articles, source code analysis, practical examples, and interview question summaries, making it invaluable for both newcomers and experienced developers preparing for competitive tech positions.

Langflow is an open-source platform for building and deploying AI-powered agents and workflows, providing developers with both a visual no-code builder interface and full Python source code access for customization. It supports all major LLMs, vector databases, and AI tools while offering deployment options as APIs, JSON exports, or MCP servers that integrate with any application framework. The platform features an interactive playground for testing flows, multi-agent orchestration, enterprise-grade security, and observability integrations, with an easy-to-use desktop application available for Windows and macOS.

Awesome OpenClaw Skills is a curated collection of 5,200+ community-built skills for OpenClaw, a locally-running AI assistant, organized by category for easy discovery and installation. The repository filters and categorizes skills from the official OpenClaw Skills Registry, removing spam, duplicates, low-quality content, and malicious entries. It serves as the primary resource for developers and users seeking to extend OpenClaw's capabilities with external service integrations, automation workflows, and specialized tasks.

Dify is an open-source, production-ready platform for building and deploying agentic AI workflows, combining a visual workflow builder, RAG pipeline, agent capabilities, and comprehensive LLM integrations. It serves developers and organizations who want to quickly go from prototype to production with support for multiple model providers (GPT, Llama, Mistral, etc.), observability tools, and both cloud and self-hosted deployment options. The platform features a low-code/no-code interface with prompt IDE, document processing, and extensive customization capabilities, making AI application development accessible to both technical and non-technical users.

Scrcpy is a lightweight, open-source application that displays and controls Android devices connected via USB or TCP/IP from a computer without requiring root access or installation on the device. It features high-performance screen mirroring (30-120fps), low latency (35-70ms), audio forwarding, recording capabilities, and support for keyboard, mouse, and gamepad input across Linux, Windows, and macOS platforms. The tool is designed for developers, enthusiasts, and anyone needing to interact with their Android device from a desktop computer with minimal overhead and maximum convenience.

World Monitor is a real-time global intelligence dashboard that aggregates AI-powered news from 500+ feeds, monitors geopolitical events, and tracks infrastructure across 45 interactive map layers. Built with TypeScript and available as a web app, native desktop application (Windows/macOS/Linux), and multiple domain-specific variants (tech, finance, commodity), it serves analysts, researchers, and organizations needing unified situational awareness with no API keys required when running locally with Ollama.

GSD (Get Shit Done) is a meta-prompting and context engineering framework designed to enhance AI-assisted development with Claude, Gemini, and other code generation models. This archived repository served as the original home for the project, which has since moved to GSD Redux in the Open GSD organization. The system provides spec-driven development capabilities, allowing developers to structure prompts and context to improve AI code generation quality and consistency.

Open WebUI is a self-hosted, user-friendly AI platform that provides a web interface for interacting with various large language models (LLMs) including Ollama and OpenAI-compatible APIs. It features advanced capabilities like retrieval-augmented generation (RAG), voice/video integration, image generation, and enterprise-grade authentication, making it ideal for organizations and individuals seeking offline-capable AI deployment. The platform is extensible through plugins and pipelines, supports multiple vector databases, and offers both community and enterprise versions with comprehensive security, scalability, and observability features.

Spec Kit is an open-source toolkit that enables Spec-Driven Development (SDD), a paradigm shift where detailed specifications become executable and directly generate working implementations rather than just guiding them. Built in Python and supporting 30+ AI coding agents, it provides a CLI tool and set of slash commands to help developers focus on product scenarios and desired outcomes while AI agents handle code generation. The toolkit includes core commands for creating project principles, writing specifications, planning technical implementations, breaking work into tasks, and executing full implementations.

This repository is a comprehensive collection of system prompts, internal tools, and AI model configurations extracted from popular AI coding assistants and tools like Cursor, Windsurf, Devin AI, Claude Code, Replit, and many others. It serves as a resource for developers and AI enthusiasts who want to understand how leading AI tools are architected and instructed, making it useful for learning, research, and developing similar systems. The repository has gained significant traction with over 138,000 stars, though it includes security warnings about the risks of exposed prompts and system prompt extraction.

LangChain is an open-source framework for building AI agents and LLM-powered applications by chaining together interoperable components and third-party integrations. It provides a standard interface for working with language models, embeddings, vector stores, and other AI tools, enabling developers to rapidly prototype and deploy production-ready applications. With extensive integrations, model interoperability, and a rich ecosystem including LangGraph for agent orchestration and LangSmith for monitoring, LangChain helps teams adapt quickly as AI technology evolves.

Nanobot is a lightweight, open-source AI agent framework written in Python that enables you to build and deploy intelligent assistants across multiple chat platforms and tools. It supports integration with leading LLM providers like OpenAI, Claude, and others, while offering practical features like memory, scheduled tasks (cron), web search, and Model Context Protocol (MCP) support. Designed for developers who want a minimal yet production-ready agent foundation, nanobot can be deployed locally or as a long-running service with minimal overhead, supporting channels like Slack, Discord, Telegram, and more.

This is the official repository for the Go programming language, an open-source language designed for building simple, reliable, and efficient software. Go is widely used for systems programming, cloud infrastructure, and backend development due to its fast compilation, built-in concurrency support, and clean syntax. The repository contains the complete source code for the Go compiler, standard library, and toolchain, with over 134,000 stars reflecting its popularity in the developer community.

Microsoft PowerToys is a comprehensive collection of over 30 utilities designed to enhance Windows productivity and customization, featuring tools like FancyZones for window management, PowerRename for batch file renaming, Color Picker, Command Palette, and many others. Built in C# with a focus on user experience, it caters to Windows power users and developers who want to streamline their workflow and customize their desktop environment. The project is actively maintained by Microsoft with strong community support, offering multiple installation methods including direct download, Microsoft Store, and WinGet.

MarkItDown is a lightweight Python utility that converts various file formats (PDFs, Word documents, PowerPoint presentations, Excel sheets, images, audio, HTML, and more) into Markdown for use with large language models and text analysis pipelines. It prioritizes preserving document structure (headings, lists, tables, links) while generating token-efficient Markdown that LLMs naturally understand, making it ideal for preparing documents for AI processing rather than human-readable conversions. The tool supports optional dependencies for specific formats, extensible plugins for custom functionality, and integration with Azure services for advanced document understanding.

30 seconds of code is a comprehensive learning platform that provides bite-sized coding articles and snippets across multiple languages and topics including JavaScript, CSS, HTML, Git, and Node.js. Built with Astro and hosted on Netlify, it helps developers of all levels quickly find practical code solutions and explanations to level up their skills. The repository contains thousands of articles organized by topic and searchable by tags, language, and description, making it an invaluable resource for continuous learning.

Claude Code is an AI-powered terminal tool that brings Claude's capabilities directly into your development workflow, enabling you to code faster by executing routine tasks, explaining complex code, and managing git workflows entirely through natural language commands. It integrates seamlessly with your terminal, IDE, and GitHub, and supports an extensible plugin ecosystem for custom commands and agents. Designed for developers who want to leverage AI assistance while maintaining full control over their codebase, it includes safety features, data protection safeguards, and a supportive developer community.

React Native is a cross-platform mobile app framework that allows developers to build native iOS and Android applications using React and JavaScript. With its "learn once, write anywhere" philosophy, it enables code reuse across platforms while providing access to native UI controls and platform-specific APIs. Developed and maintained by Meta with support from a large community, React Native offers features like hot reloading, component-based architecture, and declarative UI design to accelerate mobile app development.

Firecrawl is a web scraping and data extraction API that enables AI agents and applications to search, scrape, and interact with websites at scale. It provides industry-leading reliability with 96% web coverage, fast P95 latency of 3.4s, and LLM-ready output formats including markdown, JSON, and screenshots. The platform is available as both an open-source project and a hosted service, offering features like web search, content extraction, browser interaction, site crawling, and intelligent agent integration via MCP (Model Context Protocol).

MiroFish is a next-generation AI prediction engine powered by multi-agent swarm intelligence technology that constructs high-fidelity digital worlds from seed data to simulate future scenarios. By leveraging thousands of intelligent agents with independent personalities and long-term memory, it enables users to predict outcomes for financial forecasting, public opinion analysis, policy testing, and creative scenarios through interactive simulations. The platform is designed for decision-makers seeking risk-free policy testing and individual users exploring imaginative "what-if" scenarios, offering both serious prediction capabilities and playful simulation experiences.

Excalidraw is an open-source, collaborative virtual whiteboard application that enables users to create hand-drawn style diagrams, wireframes, and sketches with an infinite canvas. It features real-time collaboration with end-to-end encryption, offline support, and can be used both as a standalone web application at excalidraw.com or integrated as a customizable npm package into other applications. Built with TypeScript and React, it's actively maintained and widely integrated by major companies like Google Cloud, Meta, Notion, and others.

Oh My OpenAgent (OmO) is an open-source agent harness and orchestration framework that enables seamless integration with multiple AI models and code-generation services like Claude Code, OpenAI, Anthropic, and others through a unified interface. Designed for developers and AI agents who want to break free from vendor lock-in, it provides a powerful terminal UI (TUI) and plugin system for automating complex development workflows and code generation tasks. The project is built with TypeScript and actively maintained with a thriving community, offering a compelling alternative to proprietary solutions like Cursor by delivering enterprise-grade AI-assisted development capabilities without subscription fees.

This repository maintains a comprehensive, community-curated list of SaaS, PaaS, and IaaS services that offer free tiers specifically useful for developers, DevOps practitioners, and infrastructure engineers. With over 122,000 stars and contributions from 1,600+ people, it serves as a go-to resource for finding legitimate free-tier services across dozens of categories including cloud providers, CI/CD tools, databases, monitoring, and more. The list is carefully curated to include only genuine free tiers (not trials) that meet security standards, helping developers make informed decisions about which services to adopt for their projects.

Kubernetes (K8s) is an open-source container orchestration platform that automates the deployment, scaling, and management of containerized applications across clusters of machines. Built on decades of Google's production experience with Borg and community best practices, it provides essential mechanisms for container scheduling, service discovery, and workload management. Kubernetes is hosted by the Cloud Native Computing Foundation (CNCF) and is widely adopted by organizations for running microservices-based applications at scale.

Clash Verge Rev is a modern cross-platform GUI client for Clash Meta proxy management, built with Tauri and TypeScript for Windows, macOS, and Linux. It provides a feature-rich interface with configuration management, system proxy controls, TUN mode support, visual node/rule editing, and WebDAV backup capabilities. Designed for users who want a streamlined proxy experience with an intuitive, customizable interface and powerful backend performance powered by Rust and the Clash Meta kernel.

Electron is a framework that enables developers to build cross-platform desktop applications using JavaScript, HTML, and CSS, leveraging Node.js and Chromium under the hood. It is used by major applications like Visual Studio Code and provides binaries for macOS, Windows, and Linux. The repository contains the core framework source code, build infrastructure, comprehensive documentation, and tooling to help developers create native-feeling desktop apps with web technologies.

Node.js is an open-source, cross-platform JavaScript runtime environment that enables developers to execute JavaScript code outside of web browsers. It is designed for building server-side applications, command-line tools, and other non-browser environments, and is maintained by the OpenJS Foundation with support from a Technical Steering Committee and active contributor community. This repository contains the core Node.js codebase, including its engine, standard library, build infrastructure, and comprehensive test suites for continuous integration across multiple platforms.

IPTV is a comprehensive, community-driven collection of publicly available IPTV channel streams from around the world, organized in M3U playlist format for use with compatible video players. The repository provides curated playlists by country, electronic program guide (EPG) data, an API, and tools for managing and validating stream links. It's designed for anyone looking to access live TV channels globally through standard streaming protocols, with an active community maintaining the channel database and ensuring stream availability.

shadcn/ui is a collection of beautifully designed, accessible React components built with TypeScript and Tailwind CSS that developers can customize and integrate into their own projects. It serves as both a component library and a code distribution platform, allowing developers to copy components directly into their applications and modify them as needed. The project is designed to help developers build their own custom component libraries using industry-standard tools like Next.js, Radix UI, and TanStack, making it ideal for teams wanting a foundation of quality UI components they can fully control and extend.

RustDesk is an open-source remote desktop application built in Rust that provides a self-hosted alternative to TeamViewer and AnyDesk. It works cross-platform (Windows, macOS, Linux, Android, iOS) with no configuration required and gives users full control over their data and security. The project supports peer-to-peer connections, custom relay servers, and features like file transfer and chat, making it ideal for users seeking privacy-focused remote access solutions.

ComfyUI is a powerful, modular AI engine designed for visual professionals who need fine-grained control over diffusion models, parameters, and creative outputs. It provides a node-based graph interface that supports image, video, 3D model, and audio generation using state-of-the-art models like Stable Diffusion, SDXL, Flux, and many others. Available as a desktop application, cloud service, or self-hosted solution across Windows, Linux, and macOS, ComfyUI empowers creators to build complex workflows without coding while integrating seamlessly into production pipelines through its API.

Awesome Hacking is a comprehensive curated collection of awesome lists and resources for hackers, penetration testers, and security researchers. It aggregates links to specialized repositories covering diverse security domains including bug bounties, malware analysis, reverse engineering, fuzzing, OSINT, and many others. This meta-repository serves as a central hub to help security professionals discover organized learning materials and tools across virtually every aspect of cybersecurity.

This is the official source code repository for the Rust programming language, containing the compiler (rustc), standard library, and comprehensive documentation. It serves developers and organizations worldwide who need to build reliable, efficient, and safe software with memory safety guarantees and concurrent execution without data races. The repository includes essential tooling like Cargo (package manager), rustfmt (formatter), and Clippy (linter), along with extensive infrastructure for testing, CI/CD, and community contributions.

D3.js is a free, open-source JavaScript library that enables developers to create dynamic, data-driven visualizations using SVG, Canvas, and HTML. Built on low-level web standards, it offers unparalleled flexibility for authoring custom graphics and has become the foundational building block for numerous higher-level charting libraries. For over a decade, D3 has powered groundbreaking visualizations and fostered a vibrant community of data practitioners worldwide.

Awesome LLM Apps is a curated collection of 100+ ready-to-run AI agent and RAG application templates that developers can clone, customize, and deploy to production. Each template is hand-built with full source code, runs in just 3 commands, and covers the modern AI stack including single agents, multi-agent teams, voice agents, MCP agents, RAG systems, and fine-tuning examples. The repository works with popular LLM providers like Claude, Gemini, OpenAI, and Llama, and includes step-by-step tutorials on the Unwind AI platform for learning how to build production-grade AI applications.

This is a comprehensive 21-lesson course by Microsoft designed to teach beginners how to build Generative AI applications, covering topics like prompt engineering, LLMs, image generation, and semantic search. The course provides hands-on code examples in both Python and TypeScript, with support for multiple AI providers including Azure OpenAI, OpenAI API, and GitHub Models. It is translated into 50+ languages and is ideal for developers with basic programming knowledge who want to learn practical generative AI development skills.

Godot Engine is a free, open-source, cross-platform game engine for creating 2D and 3D games with a unified interface and one-click export to desktop, mobile, web, and console platforms. Built in C++ and governed by the permissive MIT license, it empowers developers to retain complete ownership of their games while benefiting from a comprehensive set of development tools and an active community-driven development process. The engine is backed by the Godot Foundation and includes extensive documentation, official demos, and community resources to support game developers at all skill levels.

Google Workspace CLI (`gws`) is a unified command-line tool that provides dynamic access to all Google Workspace APIs (Drive, Gmail, Calendar, Sheets, Docs, Chat, Admin, and more) by automatically building its command surface from Google's Discovery Service. Built in Rust and designed for both humans and AI agents, it features zero boilerplate, structured JSON output, and over 40 pre-built agent skills for seamless automation and integration.

ZeroClaw is a fast, lightweight AI personal assistant infrastructure written in Rust that runs entirely on your machine with full data and control ownership. It supports multiple LLM providers (Anthropic, OpenAI, Ollama, etc.), integrates with 30+ communication channels (Discord, Telegram, Matrix, email, etc.), and provides autonomous tool execution through shell, browser, HTTP, and hardware interfaces. Designed for developers and privacy-conscious users, it emphasizes security-first defaults with configurable autonomy levels, while remaining fully deployable across any OS or platform.

This repository provides production-ready AI agents and skills for financial services workflows built on Claude, covering investment banking, equity research, private equity, and wealth management. It offers named workflow agents (like Pitch Agent, GL Reconciler, Market Researcher) that can be deployed either as Claude Cowork plugins or through the Managed Agents API, along with reusable vertical plugin bundles and data connectors. Finance professionals can install these to automate tasks like modeling, research, reconciliation, and reporting while maintaining human oversight and compliance requirements.

Axios is a widely-used, promise-based HTTP client library for JavaScript that works seamlessly in both browser and Node.js environments. It simplifies making HTTP requests with features like interceptors, request cancellation, timeout handling, and automatic JSON transformation, making it an essential tool for developers building web applications and APIs. With over 109,000 stars on GitHub, Axios has become the de facto standard for HTTP communication in the JavaScript ecosystem.

TypeScript is Microsoft's open-source programming language that adds optional static typing to JavaScript, enabling developers to build large-scale applications with better tooling support and error detection. The language compiles to clean, readable JavaScript that runs on any browser, host, or OS, making it ideal for teams building complex JavaScript applications. With over 100,000 GitHub stars and millions of weekly npm downloads, TypeScript has become the industry standard for typed JavaScript development.

Tauri is a framework for building lightweight, high-performance desktop and mobile applications using web technologies (HTML, CSS, JavaScript) for the frontend and Rust for the backend. It leverages native webview components across platforms (WKWebView on macOS/iOS, WebView2 on Windows, WebKitGTK on Linux, and Android System WebView) to deliver minimal binary sizes and superior security. The framework includes built-in tools for app bundling, auto-updates, system tray integration, notifications, and is ideal for developers who want to combine the flexibility of web development with the power and efficiency of Rust.

v2rayN is a cross-platform GUI client for Windows, Linux, and macOS that enables users to manage and connect through various proxy protocols including Xray, sing-box, V2Ray, Shadowsocks, Trojan, and others. Built in C#, it provides a user-friendly interface for configuring complex proxy settings, managing subscriptions, testing server speeds, and routing rules. The project is actively maintained with over 107k stars and includes comprehensive support for multiple protocols and configuration formats.

Claude-Mem is a persistent memory compression system that enables AI agents like Claude Code, Gemini CLI, and other LLM-powered tools to maintain context across multiple sessions. It automatically captures agent activities, compresses them using AI, stores them in a searchable vector database, and intelligently injects relevant context into future sessions. The system is designed for developers and teams working with AI agents who need continuity of knowledge about their projects without manual context management.

OpenHuman is an open-source personal AI assistant built with Rust that combines local privacy with managed cloud services, enabling users to integrate AI into their daily workflows. It features a clean desktop UI, local memory management, and customizable integrations while maintaining user control over sensitive data. The project is actively developed, cross-platform compatible (macOS, Linux, Windows), and designed to be both simple and powerful for everyday AI assistance.

frp is a high-performance reverse proxy written in Go that enables users to securely expose local servers behind NAT or firewalls to the internet. It supports TCP, UDP, HTTP, and HTTPS protocols with features including P2P mode, encryption, compression, load balancing, and dynamic proxy management. The project is actively maintained with extensive configuration options and is designed for developers and system administrators who need reliable intranet penetration and service tunneling capabilities.

Deno is a modern runtime for JavaScript, TypeScript, and WebAssembly built on V8, Rust, and Tokio that emphasizes secure defaults and excellent developer experience. It provides a complete alternative to Node.js with built-in tooling for testing, linting, and formatting, making it suitable for developers building web servers, CLIs, and other JavaScript/TypeScript applications. The project is designed to address Node.js limitations through improved security, simpler dependency management, and better TypeScript support out of the box.

Papers We Love is a community-driven repository that curates and organizes academic computer science papers from across the web, making landmark research accessible to practitioners and students. The project serves as a central directory of foundational and contemporary papers spanning topics like distributed systems, databases, algorithms, and artificial intelligence, complemented by local chapter meetups and online discussions. While respecting copyright licenses, the repository provides links to papers and facilitates learning through community engagement on Discord, meetups, and presentations. This resource is invaluable for anyone seeking to deepen their understanding of computer science theory and practice through peer-reviewed academic work.

Antigravity Awesome Skills is a searchable, installable library of 1,480+ reusable AI agent skills (playbooks) designed for Claude Code, Cursor, Codex CLI, Gemini CLI, and other AI coding assistants. It provides an npm-based installer, curated bundles, workflows, and practical documentation to help developers quickly integrate structured prompts and operating instructions into their AI tools. The repository serves both individual developers and teams seeking to standardize agent behavior across coding, testing, security, infrastructure, and product workflows.

Awesome Mac is a comprehensive, community-curated collection of high-quality macOS software and applications organized into easy-to-browse categories. Designed for macOS users, developers, and designers, it provides a searchable resource for discovering the best tools, apps, and utilities available for the platform. The project maintains multilingual support and publishes curated data in JSON format for programmatic access, making it both a reference guide and a data source for the macOS ecosystem.

Gemini CLI is an open-source terminal application that integrates Google's Gemini AI models directly into the command line, enabling developers to leverage advanced AI capabilities for code understanding, generation, automation, and task execution. Built with TypeScript and supporting multiple authentication methods (OAuth, API keys, Vertex AI), it features built-in tools like Google Search grounding, file operations, and extensibility through the Model Context Protocol (MCP). The tool is designed for developers who prefer terminal-based workflows and offers both interactive and scripting modes with free tier access and support for the latest Gemini 3 models.

This is Microsoft's official repository for Windows Terminal, a modern, feature-rich terminal application for Windows command-line users, along with the original Windows console host and shared components. The project provides a productive, tabbed terminal experience with rich text support, extensive customization options, and integration with Windows Subsystem for Linux (WSL), serving millions of developers and command-line users. It is actively maintained, welcomes community contributions, and is available through multiple installation methods including the Microsoft Store, GitHub releases, and package managers like winget and Chocolatey.

This is an AI-powered stock analysis system that automatically analyzes A-shares, Hong Kong stocks, and US stocks using large language models, aggregating real-time market data, news, and technical indicators. It generates daily intelligent decision dashboards with AI insights, risk alerts, and trading recommendations, which are pushed to multiple channels like WeChat, Feishu, Telegram, Discord, Slack, and email. The system runs zero-cost on GitHub Actions, local servers, or Docker, making it accessible to retail investors and traders who want AI-driven stock analysis without server costs.

Supabase is an open-source Firebase alternative that provides a complete backend platform built on PostgreSQL, offering hosted databases, authentication, auto-generated APIs (REST, GraphQL, Realtime), serverless functions, file storage, and AI/vector toolkit. It's designed for developers building web, mobile, and AI applications who want enterprise-grade, self-hosted or managed solutions with the developer experience of Firebase but using open-source tools. The repository contains the full Supabase ecosystem including the dashboard, client libraries, and supporting services like PostgREST, GoTrue, and Realtime.

CC Switch is a cross-platform desktop application built with Rust and Tauri that serves as an all-in-one manager for multiple AI coding assistants including Claude Code, Codex, Gemini CLI, OpenCode, OpenClaw, and Hermes Agent. It enables users to seamlessly switch between different AI providers, manage API keys, configure proxy settings, and handle advanced features like MCP servers, skills management, and provider routing. Designed for developers and AI enthusiasts, CC Switch simplifies the complexity of managing multiple AI coding tools through an intuitive interface with support for Windows, macOS, and Linux.

Immich is a high-performance, self-hosted photo and video management solution designed as a privacy-focused alternative to Google Photos. It offers both mobile and web applications with features like automatic backup, facial recognition, advanced search, multi-user support, and album sharing, enabling users to maintain complete control over their media library. Built with TypeScript, NestJS, Flutter, and Svelte, it serves individuals and families seeking a feature-rich, self-hosted photo management platform.

This repository is a curated collection of 1000+ production-ready Claude Skills and plugins that extend Claude AI's capabilities beyond text generation to perform real-world actions like sending emails, creating issues, and automating workflows across 500+ integrated applications. It serves developers and teams looking to enhance Claude's productivity across various domains including document processing, development tools, data analysis, business automation, and more. The repository includes both community-contributed and official skills organized by use case, complete with installation instructions and integration guides powered by Composio's app automation platform.

CodeWhale is a terminal-based AI coding agent powered by DeepSeek V4 that runs from the command line, streams reasoning blocks, and edits local workspaces with built-in approval gates. Built in Rust as a matched pair of self-contained binaries (`codewhale` dispatcher and `codewhale-tui` TUI runtime), it features a sophisticated "Constitution" system that establishes clear hierarchies of authority for decision-making, concurrent sub-agents, LSP integration for live diagnostics, and an auto-routing mode that intelligently selects the model and thinking level per turn. The tool is designed for developers who need a trustworthy, transparent agentic coding assistant that prioritizes verification, user intent, and workspace legibility.

HowToCook is a community-driven, open-source cookbook designed specifically for programmers who want to cook at home, featuring clear and precise recipe instructions in Simplified Chinese. The repository contains hundreds of recipes organized by dish type (vegetarian, meat, seafood, breakfast, desserts, drinks, and condiments), along with cooking tips and techniques, accessible both as markdown files and through an interactive web interface. It addresses the frustration programmers often have with poorly-formatted traditional recipes by presenting ingredients and instructions in a structured, formal manner that's easy to follow. The project welcomes contributions from the community to continuously expand and improve its collection of recipes.

PyTorch is a widely-used Python library that provides tensor computation with strong GPU acceleration and a tape-based automatic differentiation system for building dynamic neural networks. It combines the flexibility of NumPy-like tensor operations with the power of deep learning frameworks, making it ideal for both GPU-accelerated scientific computing and cutting-edge machine learning research. PyTorch is designed to be intuitive and Pythonic, allowing researchers and developers to build and train neural networks with maximum flexibility and minimal overhead.

Angular is a comprehensive development platform built with TypeScript for creating modern mobile and desktop web applications with confidence. It provides developers with powerful tools, comprehensive documentation, and an excellent developer experience through features like the Angular CLI, reactive components, forms management, and server-side rendering capabilities. The repository serves as the source code for the Angular framework and its ecosystem, making it an essential resource for millions of developers building scalable and high-performance web applications.

Neovim is a modern refactor of Vim that prioritizes extensibility, usability, and maintainability through a cleaner codebase and robust API. It supports multiple programming languages via its RPC API, includes advanced features like asynchronous job control and embedded terminal emulation, and maintains compatibility with most Vim plugins. Designed for both experienced Vim users and developers, Neovim enables creation of custom UIs and scripting in Lua while remaining a highly efficient text editor.

Taste-Skill is an anti-slop frontend framework that provides portable AI agent skills to help Claude, Codex, and other AI coding assistants generate high-quality, polished user interfaces instead of generic, boilerplate designs. It includes implementation skills for code generation, image-generation skills for design reference boards, and specialized variants (minimalist, brutalist, soft design) to match different visual directions. Developers can install individual skills via the npx CLI or use them directly in ChatGPT and coding agents to improve layout, typography, motion, and spacing in AI-generated frontends.

FastAPI is a modern, high-performance Python web framework for building APIs with automatic OpenAPI documentation and built-in data validation through Pydantic. Designed for rapid development with intuitive type hints support, it enables developers to create production-ready REST APIs with less code and fewer bugs while maintaining performance comparable to Node.js and Go. The framework is widely used by companies like Microsoft, Uber, and Netflix for building scalable, well-documented API services.

Material UI is a comprehensive React component library that implements Google's Material Design system, trusted by major product teams worldwide and battle-tested through over a decade of open-source development. It provides a rich set of pre-built, customizable components that enable developers to build consistent, visually appealing user interfaces following Material Design principles. The library is free, MIT-licensed, and can be extended with MUI X for more advanced use cases.

Ant Design is an enterprise-class UI design language and React component library that provides a comprehensive set of high-quality, production-ready components built with TypeScript. Designed for developers building web applications that require professional, accessible interfaces, it offers powerful theme customization, internationalization support for dozens of languages, and follows modern design principles. The library is widely adopted across the industry, with nearly 100k GitHub stars, and includes extensive documentation, design resources, and development tools to accelerate UI development.

This repository is a comprehensive guide and reference implementation for mastering Claude Code, Anthropic's AI-powered development environment, covering best practices for agents, commands, skills, workflows, and advanced features. It provides practical examples, configuration templates, and documentation for developers looking to leverage agentic engineering patterns and move beyond basic "vibe coding" to structured, production-ready AI workflows. The project has gained significant community attention (55k+ stars) and serves as both an educational resource and a working codebase demonstrating real-world implementations of Claude Code capabilities.

This repository provides a comprehensive implementation of a ChatGPT-like large language model (LLM) built from scratch using PyTorch, accompanying the book "Build a Large Language Model (From Scratch)." It guides learners through each stage of LLM development—from text tokenization and attention mechanisms to model architecture, pretraining, and fine-tuning—with both Jupyter notebooks and Python scripts that mirror the techniques used in production models like ChatGPT. The repository is designed for students, researchers, and practitioners who want to understand how LLMs work by building them step-by-step, with additional bonus materials covering advanced optimization techniques, alternative architectures, and implementations of popular models like Llama and Qwen.

Browser Use is a Python framework that enables AI agents to automate web tasks by controlling browsers programmatically. It provides a high-level API for building autonomous agents that can fill forms, shop online, search the web, and perform complex multi-step browser interactions. The framework supports multiple LLM providers (OpenAI, Anthropic, Google, etc.) and offers both open-source and cloud-hosted deployment options with advanced features like stealth browsing, proxy rotation, and captcha solving.

This repository is Microsoft's comprehensive 12-week web development curriculum designed for absolute beginners, featuring 24 lessons that cover HTML, CSS, and JavaScript through hands-on projects like terrariums, browser extensions, and games. The course is available in over 50 languages and includes quizzes, assignments, and challenges to help learners build practical skills while understanding fundamental web development concepts. It's ideal for self-paced learners or educators looking for an open-source, project-based curriculum to teach web development from the ground up.

Tailwind CSS is a utility-first CSS framework that enables developers to rapidly build custom user interfaces by composing pre-defined utility classes directly in HTML. Written primarily in TypeScript with Rust-based performance optimizations, it provides a highly configurable and responsive design system that prioritizes functional CSS over traditional stylesheets. The project is actively maintained with comprehensive documentation, a supportive community, and extensive testing across multiple build tool integrations including PostCSS, Vite, Webpack, and Next.js.

llmfit is a terminal tool that helps users find which large language models will run efficiently on their specific hardware by detecting system specs and scoring models across quality, speed, memory fit, and context dimensions. Built in Rust with an interactive TUI interface, it supports hundreds of models and multiple providers (Ollama, llama.cpp, MLX, Docker) while offering features like hardware simulation, performance benchmarking, and a community leaderboard for real-world performance metrics. The tool is designed for anyone running local LLMs who wants to make informed decisions about which models suit their hardware before installation.

Puppeteer is a JavaScript library that provides a high-level API to automate and control Chrome and Firefox browsers through the DevTools Protocol or WebDriver BiDi. It enables developers to perform headless browser automation for tasks like testing, web scraping, performance testing, and automated form submission. With over 94,000 stars, Puppeteer is widely used for end-to-end testing and browser automation in Node.js environments.

Oh My codeX (OMX) is a multi-agent orchestration layer and workflow engine built on top of OpenAI's Codex CLI that enhances developer productivity with 30+ specialized agent prompts, 35+ workflow skills, and structured planning features. It provides a complete workflow from clarification to execution through tools like `$deep-interview`, `$ralplan`, and `$ultragoal`, while maintaining project state and guidance in a `.omx/` directory. OMX is designed primarily for macOS/Linux users who want a more powerful and organized development experience with Codex, featuring team coordination capabilities, research modes, and durable goal management.

This comprehensive repository provides well-documented implementations of classic and modern design patterns in Java, serving as both a learning resource and practical reference for developers and architects. With over 94,000 stars and contributions from hundreds of developers, it covers creational, behavioral, and structural patterns with clean, tested code examples and detailed explanations. The project is designed for Java developers of all skill levels who want to understand and apply proven design patterns to improve their software architecture and maintainability.

PicoClaw is an ultra-lightweight AI assistant written in Go that runs on minimal hardware ($10 devices with <10MB RAM) while maintaining fast performance and broad compatibility across ARM, RISC-V, MIPS, and x86 architectures. Designed for users who need AI capabilities on edge devices or low-power systems, it supports multiple LLM providers, communication channels (Discord, WeChat, Telegram, etc.), and the Model Context Protocol for extensibility. The project was notably bootstrapped through AI-assisted development, with the majority of its core code generated and refined by AI agents working with human oversight.

NVM (Node Version Manager) is a POSIX-compliant bash script that allows developers to quickly install, manage, and switch between multiple active Node.js versions on a per-user, per-shell basis. It works across Unix, macOS, and Windows WSL, supporting various shells including bash, zsh, and fish, making it essential for developers who need to work with different Node.js versions simultaneously. The tool is highly popular with over 93,500 stars and provides features like version aliases, `.nvmrc` file support, and automatic version switching based on project requirements.

Deep-Live-Cam is a real-time deepfake and face-swap application that enables users to swap faces in live video streams with just a single reference image and a few clicks. Built with Python and powered by AI models like GFPGAN and InsightFace, it supports GPU acceleration (NVIDIA, AMD, Apple Silicon) for fast processing and includes features like face mapping, mouth masking, and content safety checks. The tool is designed for creative applications such as character animation, content creation, live performances, and entertainment, while emphasizing ethical use and requiring user consent when using real people's faces.

Bun is an all-in-one JavaScript and TypeScript toolkit that serves as a fast, drop-in replacement for Node.js, combining a runtime, bundler, test runner, and package manager into a single executable. Written in Zig and powered by JavaScriptCore, Bun dramatically reduces startup times and memory usage while supporting TypeScript and JSX out-of-the-box. It's designed for developers building JavaScript/TypeScript applications who want significantly faster performance and a unified toolchain without the overhead of multiple separate tools.

Free Claude Code is a proxy middleware that routes Anthropic API traffic from Claude Code to multiple LLM providers (NVIDIA NIM, OpenRouter, Google Gemini, DeepSeek, Mistral, Groq, Ollama, and others), enabling users to access Claude Code functionality with free, paid, or local models. It supports CLI, VS Code, JetBrains IDEs, and Discord/Telegram integrations, with features like streaming, tool use, voice transcription, and per-model routing through an intuitive admin UI. Designed for developers who want to use Claude Code's capabilities without being locked into Anthropic's API, it maintains protocol compatibility while offering flexibility in model selection and provider choice.

This repository is a weekly newsletter for technology enthusiasts called "科技爱好者周刊" (Tech Enthusiast Weekly), published every Friday by Ruan Yifeng. It curates and shares noteworthy technology news, articles, software, and resources, with over 398 issues archived since its inception. The project also maintains a job board section where programmers can post and discover employment opportunities, making it a valuable resource for both tech knowledge and career development in the programming community.

Oh-My-ClaudeCode is a multi-agent orchestration system for Claude Code that enables teams-first agentic development with zero learning curve. It provides specialized AI agents (executor, debugger, architect, etc.) that work together to automate coding tasks, code review, debugging, and project management through simple natural-language commands like `/autopilot` and `/team`. Designed for developers who want powerful AI-assisted development without mastering complex agent coordination, it features parallel execution, intelligent context management, and seamless integration with Claude Code's native workflow.

agent-browser is a high-performance browser automation CLI built in Rust, designed specifically for AI agents to interact with web applications programmatically. It provides a comprehensive set of commands for browser control including navigation, element interaction, screenshots, and DOM inspection, with support for both reference-based and traditional selectors optimized for AI workflows. The tool is installable via npm, Homebrew, or Cargo, and includes an AI chat interface for natural language browser control, making it ideal for developers building autonomous agent applications.

NanoClaw is a lightweight, secure AI assistant framework that runs Claude agents in isolated Docker containers for personal use. It connects to multiple messaging platforms (WhatsApp, Telegram, Discord, Slack, and others) with built-in memory, scheduled task support, and native integration with Anthropic's Claude Agent SDK. Designed as a simpler, more understandable alternative to OpenClaw, NanoClaw prioritizes security through OS-level container isolation, customizability through code modification rather than configuration files, and transparency with a minimal codebase that users can fully understand and extend.

Multica is an open-source managed agents platform that transforms AI coding agents into autonomous teammates. It enables users to assign tasks to agents, track progress, and build reusable skills across teams—similar to managing human colleagues. The platform supports multiple agent providers (Claude Code, Codex, GitHub Copilot CLI, and others) and offers features like task assignment, autonomous execution, squads for team routing, autopilots for scheduled work, and unified runtime management.

Storybook is the industry-standard frontend workshop that enables developers to build, document, and test UI components in isolation across multiple frameworks including React, Vue, Angular, Web Components, and more. With over 90,000 stars, it provides a comprehensive platform with extensive addon ecosystem, interactive documentation capabilities, and testing tools that help thousands of teams improve UI development workflows. The repository serves developers and design systems teams who need a collaborative environment to create bulletproof, well-documented UI components.

Playwright is a comprehensive framework for web testing and automation that enables developers and AI agents to control Chromium, Firefox, and WebKit browsers through a single unified API. It offers multiple interfaces including a full-featured test runner (Playwright Test), a CLI for coding agents, an MCP server for AI integration, and a library for browser automation scripts. Useful for end-to-end testing, web scraping, screenshot capture, and any workflow requiring programmatic browser control with features like auto-waiting, web-first assertions, test isolation, and cross-browser parallelism.

CodeGraph is a pre-indexed semantic code knowledge graph system designed to supercharge AI coding agents (Claude Code, Cursor, Codex, and others) by providing instant symbol lookup, call graphs, and code structure analysis without expensive file scanning. It reduces API costs by ~35%, cuts token usage by ~70%, and minimizes tool calls by ~71% while running 100% locally, making it ideal for developers using AI-powered IDEs who want faster and cheaper code exploration.

Bitcoin Core is the reference implementation of the Bitcoin protocol, a peer-to-peer cryptocurrency network client written in C++. It enables users to download and validate the complete blockchain, manage transactions, and optionally use an integrated wallet and graphical interface. The repository serves as the official staging tree for Bitcoin Core development, with rigorous testing, code review, and multi-platform CI/CD pipelines to ensure security and stability for this critical financial software.

Gin is a high-performance HTTP web framework written in Go that provides a fast, Express.js-like API for building REST APIs, web applications, and microservices. It offers up to 40 times better performance than comparable frameworks through its efficient httprouter implementation, while maintaining simplicity and ease of use. Gin is ideal for developers who need to build scalable, high-throughput services with minimal boilerplate and excellent developer productivity.

Mermaid is a JavaScript-based tool that enables users to create diagrams—such as flowcharts, sequence diagrams, class diagrams, and more—using a simple Markdown-like syntax. It solves the problem of documentation falling out of sync with development by making diagrams easy to create, modify, and integrate into code repositories and documentation workflows. The tool is accessible to both developers and non-developers through its live editor and integrations with popular platforms like GitHub, making it a valuable asset for teams that want to maintain current, clear technical documentation.

Hugo is a fast and flexible static site generator written in Go that enables developers to build websites in seconds with advanced templating, asset processing, and multilingual support. It serves diverse use cases including blogs, documentation sites, portfolios, and corporate websites, and features powerful tools for CSS/Sass/Tailwind processing, image optimization, and JavaScript bundling. Hugo is optimized for speed and designed to work seamlessly with Git-based deployment workflows, making it ideal for developers and content creators who want a modern, efficient site-building experience.

NextChat is a lightweight, cross-platform AI chat application that supports multiple LLM providers including ChatGPT, Claude, Gemini, and DeepSeek. Built with Next.js and React, it offers seamless deployment options (web, desktop, iOS, Android) with privacy-first design, markdown rendering, and advanced features like prompt templates, plugins, and real-time chat capabilities. The application is designed for users who want fast, feature-rich AI interactions across any device with a focus on data privacy and customization.

This repository is a comprehensive, community-curated directory of Model Context Protocol (MCP) servers that enable AI models to securely interact with local and remote resources. It serves developers and AI enthusiasts looking to extend AI capabilities through standardized server implementations across dozens of categories including databases, APIs, file systems, and cloud platforms. The repository features a well-organized list with multilingual support, community resources, and links to a web-based directory for easy discovery and integration of MCP servers.

OpenCV is a leading open-source computer vision library written in C++ that provides comprehensive tools for image processing, video analysis, object detection, and deep learning applications. Designed for developers and researchers, it offers cross-platform support, extensive documentation, and a large community ecosystem, making it the standard choice for building real-world computer vision solutions across industries. The library includes optimized algorithms for performance-critical tasks and integrates with popular deep learning frameworks.

Django is a high-level Python web framework designed for rapid development and pragmatic, clean design, enabling developers to build web applications with batteries-included features like an ORM, admin interface, and template system. It is the go-to framework for perfectionists with deadlines, offering comprehensive documentation, a vibrant community, and extensive testing infrastructure. Django supports modern Python versions (3.12+) and includes built-in tools for deployment, security, and scalability, making it suitable for everything from small projects to large enterprise applications.

Home Assistant is an open-source home automation platform that prioritizes local control and privacy by running on local hardware like Raspberry Pi or personal servers. Built with Python and designed for a community of DIY enthusiasts, it uses a modular architecture to easily integrate hundreds of smart home devices and services. The platform provides a web interface for managing automations, device controls, and integrations while keeping all data and processing local rather than relying on cloud services.

Uptime Kuma is a self-hosted monitoring tool that tracks the availability and performance of websites, APIs, and services with an intuitive, real-time web interface. It supports monitoring via multiple protocols (HTTP, TCP, DNS, WebSocket, Ping, etc.) and can send notifications through 90+ services including Telegram, Discord, and Slack. Built with Vue 3 and Node.js, it's designed for easy deployment via Docker while being lightweight enough for self-hosting on personal infrastructure.

nanochat is a minimal, hackable experimental framework for training large language models on a single GPU node, covering the entire pipeline from tokenization and pretraining to finetuning, evaluation, and inference. It enables users to train GPT-2-level capability models for under $100 in just a few hours and interact with them through a ChatGPT-like web interface, with automatic hyperparameter optimization controlled by a single complexity dial. The repository maintains a competitive leaderboard for "GPT-2 speedruns" and is designed to be both beginner-friendly and suitable for LLM researchers conducting scaling experiments.

Codex CLI is a lightweight AI-powered coding agent from OpenAI that runs locally in your terminal, enabling developers to leverage advanced code generation and assistance capabilities without leaving their command line. Built in Rust for performance and reliability, it integrates with ChatGPT plans and can also be used via API keys, providing a seamless development experience across macOS, Linux, and Windows. The tool is designed for developers who want AI-assisted coding directly in their terminal workflow, complementing IDE extensions and the web-based Codex experience.

Svelte is a modern JavaScript compiler that transforms declarative components into highly efficient JavaScript code that surgically updates the DOM, offering a fundamentally different approach to building web applications. Designed for developers who want better performance and simpler code, Svelte eliminates the runtime overhead of traditional frameworks by compiling components at build time. This monorepo contains the core Svelte compiler, packages, documentation, and tooling, supported by an active community and comprehensive testing infrastructure.

This repository contains reference implementations of Model Context Protocol (MCP) servers built by the MCP steering group to demonstrate how to give Large Language Models secure, controlled access to tools and data sources. It includes educational examples such as servers for file system operations, Git repositories, web fetching, memory systems, and time utilities, all designed to showcase MCP features and SDK usage across multiple programming languages. These servers serve as templates for developers building their own MCP implementations rather than production-ready solutions.

ML-For-Beginners is a comprehensive 12-week curriculum created by Microsoft that teaches classic machine learning fundamentals through 26 lessons with hands-on projects, quizzes, and assignments. Designed for beginners, it uses Python with Scikit-learn and covers regression, classification, clustering, and web applications, complemented by real-world datasets from diverse global cultures. The course is freely available, open-source, translated into 50+ languages, and emphasizes learning-by-building pedagogy to help skills stick.

uv is an extremely fast Python package and project manager written in Rust that serves as a single unified tool to replace pip, pip-tools, pipx, poetry, pyenv, and other Python development utilities. It offers 10-100x faster performance than traditional tools while providing comprehensive features including project management with lockfiles, dependency resolution, Python version management, script execution with inline dependencies, and tool installation. Designed for Python developers of all experience levels, uv is cross-platform, installable without requiring Rust or Python, and backed by Astral, the creators of Ruff.

Syncthing is an open-source, peer-to-peer file synchronization application written in Go that continuously syncs files between two or more computers while prioritizing data safety, security, and ease of use. It runs on Windows, macOS, Linux, and other platforms, offering both CLI and GUI interfaces, with features like encrypted transfers, automatic syncing, and cross-platform compatibility. The project is designed for individual users who want decentralized, secure file synchronization without relying on cloud services.

Laravel is a popular PHP web application framework that provides an elegant, expressive syntax for building robust and scalable web applications. It simplifies common development tasks like routing, dependency injection, database management with ORM (Eloquent), migrations, job queuing, and real-time event broadcasting. This repository serves as the starter skeleton application for developers to quickly begin building Laravel projects with a pre-configured directory structure and essential dependencies. The framework is designed to be accessible to beginners while powerful enough for large enterprise applications, and it includes comprehensive documentation and AI-friendly conventions for modern development workflows.

This repository provides a comprehensive collection of AI agent skills for marketing tasks, enabling Claude Code, OpenAI Codex, and other AI agents to assist with conversion optimization, copywriting, SEO, analytics, and growth engineering. Designed for technical marketers and founders, each skill is a markdown-based workflow that teaches AI agents specialized knowledge and best practices for specific marketing domains, with skills cross-referencing each other to build on shared context. The repository is highly modular with 40+ individual skills covering everything from CRO and content strategy to paid advertising, email marketing, and sales enablement, making it a practical toolkit for automating marketing work at scale.

Sherlock is a Python-based OSINT tool that hunts down social media accounts by username across 400+ social networks, making it invaluable for security researchers, penetrators, and investigators. It offers a command-line interface with flexible output options (text, CSV, XLSX, JSON), proxy support, and batch processing capabilities for efficient username reconnaissance. The tool is actively maintained, well-documented, and available through multiple installation methods including pip, Docker, and package managers.

Zed is a high-performance, multiplayer code editor built in Rust by the creators of Atom and Tree-sitter, designed for collaborative development with a focus on speed and responsiveness. The project is actively developed with a large community, offering native support for macOS, Linux, and Windows, along with extensive features like AI integration, debugging support, and an extension system. It's ideal for developers seeking a modern alternative to traditional code editors with built-in collaboration capabilities and strong performance characteristics.

Mall is a comprehensive e-commerce system built with Spring Boot and MyBatis that includes both a customer-facing storefront and a complete backend management system. It features product management, order processing, member services, promotional campaigns, and content management, with deployment via Docker and integration of modern technologies like Elasticsearch, Redis, RabbitMQ, and MongoDB. This project serves as a production-ready reference implementation for developers learning full-stack e-commerce development, supported by extensive documentation and tutorials.

RealWorld is a comprehensive demo application collection that implements a Medium.com clone using over 100 different frontend and backend technology stacks, allowing developers to see how the same app is built with various frameworks like React, Angular, Node, Django, and many others. All implementations adhere to the same standardized API specification, making it possible to mix and match any frontend with any backend. Unlike typical "todo" demos, RealWorld strikes a sweet spot between simplicity and real-world complexity, providing developers with practical knowledge about building production-quality applications with their chosen frameworks.

This repository provides a collection of agent skills for Obsidian that enable AI agents like Claude Code and Codex CLI to interact with Obsidian vaults and create content in various formats. The skills cover Obsidian Flavored Markdown, Obsidian Bases, JSON Canvas files, and CLI operations, following the standard Agent Skills specification. It's designed for developers and knowledge workers who want to automate Obsidian workflows through AI agents.

LocalSend is a free, open-source cross-platform application built with Flutter and Dart that enables secure file and message sharing between nearby devices over a local network without requiring internet or third-party servers. It uses REST API and HTTPS encryption to provide fast, reliable local communication and is available on Windows, macOS, Linux, Android, iOS, and Fire OS. The project is highly active with 82,000+ stars and welcomes community contributions including translations and bug fixes.

Fira Code is a free, open-source monospaced font designed specifically for programmers that includes ligatures for common multi-character programming operators and symbols like `->`, `<=`, and `:=`. The font enhances code readability by rendering these character sequences as single visual units while maintaining ASCII compatibility, and it offers extensive customization through stylistic sets and character variants. With over 81,000 stars, it's widely supported across major code editors and terminals, making it an essential tool for developers who want to improve their coding experience.

This repository is a comprehensive curated list of university-level computer science courses that include video lectures, covering topics ranging from foundational programming to advanced machine learning, systems, and specialized domains. It serves as a free educational resource for students and professionals looking to access high-quality course materials from institutions like MIT, Stanford, Berkeley, and other leading universities. The repository is maintained with strict contribution guidelines to ensure only legitimate college-level courses are included, making it a reliable reference for self-directed learning in computer science.

RAGFlow is an open-source Retrieval-Augmented Generation (RAG) engine that combines advanced RAG technology with AI agent capabilities to create a superior context layer for large language models. It provides intelligent document processing, template-based chunking, reduced hallucinations through grounded citations, and support for diverse data sources, enabling developers to build production-ready AI systems with exceptional efficiency. The platform includes both cloud-hosted and self-hosted options, with pre-built agent templates and an intuitive API for seamless enterprise integration.

vLLM is a high-throughput and memory-efficient inference and serving engine for large language models, designed to make LLM serving fast, easy, and accessible. Originally developed at UC Berkeley, it features state-of-the-art optimizations like PagedAttention, continuous batching, and support for 200+ model architectures across multiple hardware platforms including NVIDIA GPUs, AMD GPUs, TPUs, and CPUs. The library provides seamless integration with Hugging Face models, OpenAI-compatible APIs, and supports advanced features like quantization, distributed inference, and multi-modal models, making it ideal for both researchers and production deployments.

Vite is a next-generation build tool and dev server designed to provide a faster and leaner development experience for modern web projects. It features instant server start, lightning-fast Hot Module Replacement (HMR), and optimized production builds using Rolldown, along with a highly extensible plugin API. Vite is ideal for frontend developers and teams building contemporary web applications who want significantly improved build performance and developer experience.

Spring Boot is a framework that simplifies the creation of production-grade, Spring-powered Java applications with minimal configuration and setup effort. It provides an opinionated approach to Spring development, eliminating the need for XML configuration or code generation while offering built-in features like embedded servers, security, metrics, and health checks. Developers can quickly build stand-alone applications using `java -jar` or traditional WAR deployments through an intuitive API and comprehensive tooling. Spring Boot is ideal for both beginners seeking a fast getting-started experience and experienced developers who need flexibility as requirements evolve.

fzf is a fast, portable command-line fuzzy finder written in Go that enables interactive selection from large datasets with fuzzy matching capabilities. It serves developers and power users who want to enhance their shell workflows with intelligent search and selection functionality, offering seamless integrations with Bash, Zsh, Fish, Vim, and Neovim. The tool is highly customizable, optimized for performance with SIMD algorithms, and provides an event-driven architecture for building rich terminal user interfaces and custom workflows.

TradingAgents is a multi-agent LLM financial trading framework that simulates real-world trading firms by deploying specialized AI agents—including fundamental analysts, sentiment analysts, technical analysts, traders, and risk managers—that collaboratively evaluate market conditions and make informed trading decisions. The framework uses multiple LLM providers (OpenAI, Anthropic, Google, DeepSeek, etc.) and integrates real-time market data to enable sophisticated, multi-perspective analysis and dynamic agent discussions. Designed for research purposes, it provides both a command-line interface and Python package for building and backtesting AI-driven trading strategies.

This repository is a comprehensive collection of AI-powered writing prompts and tools designed to help researchers and academics improve their paper writing without tedious manual polishing. It provides battle-tested prompt templates for translation, paraphrasing, and editing in both English and Chinese contexts, along with emerging agent skills that lower the barrier to entry for advanced AI-assisted writing techniques. Created by gathering best practices from top research institutions like MSRA and top-tier universities, this resource aims to democratize access to high-quality writing optimization that was previously confined to elite research groups.

Stirling PDF is a powerful, open-source PDF editing platform that enables users to edit, sign, redact, convert, and automate PDFs without relying on external services. It offers 50+ PDF tools and can be deployed as a desktop application, web browser interface, or self-hosted server with private APIs, making it ideal for individuals and enterprises seeking privacy-focused document processing. The platform features enterprise-grade capabilities including SSO, auditing, workflow automation, and REST APIs for integration, with support for 40+ languages and a large, active community.

Hoppscotch is an open-source API development ecosystem that serves as a modern alternative to Postman and Insomnia, offering a lightweight and fast web-based API client for testing and developing REST, GraphQL, WebSocket, and other API types. It features a progressive web app (PWA) that works offline, supports collections and team collaboration, includes pre/post-request scripts and testing capabilities, and is available as a web app, desktop application, and CLI tool. Developers and API teams use Hoppscotch to streamline request creation, testing, and documentation workflows with an intuitive interface and powerful development features.

Mole is a comprehensive macOS terminal utility that combines system cleaning, app uninstallation, disk analysis, and performance monitoring into a single CLI tool. It functions as an all-in-one replacement for commercial tools like CleanMyMac, AppCleaner, and DaisyDisk, helping users reclaim storage space and optimize their Mac's performance through deep cleaning of caches, logs, app remnants, and system artifacts. The tool prioritizes safety with dry-run previews, operation logging, and conservative deletion boundaries, while offering an interactive menu interface with both destructive and non-destructive analysis modes.

Netdata is an open-source, real-time infrastructure monitoring platform that provides per-second metrics collection and visualization with AI-powered anomaly detection and zero-configuration setup. Designed for DevOps teams and system administrators, it offers edge-based processing, high energy efficiency, and complete visibility across infrastructure and applications without the complexity of traditional monitoring solutions. The platform supports multiple operating systems and integrates seamlessly with popular tools like Kubernetes, Docker, Prometheus, and Grafana.

PaddleOCR is a leading open-source OCR and document AI toolkit that converts PDFs and images into structured, LLM-ready data (JSON/Markdown) with support for 100+ languages. Built on the PaddlePaddle framework, it features state-of-the-art models like PaddleOCR-VL and PP-OCRv5, offering both lightweight efficiency and production-grade accuracy for document parsing, text recognition, and knowledge information extraction. The toolkit is widely adopted by major AI projects like Dify, RAGFlow, and Cherry Studio, and provides seamless integration with various hardware backends including GPUs, CPUs, and specialized AI accelerators.

This repository is a comprehensive study guide for experienced Java backend engineers covering advanced topics essential for internet-scale systems, including high concurrency, distributed systems, high availability, microservices, and massive data processing. It provides detailed explanations of complex technologies like message queues (Kafka, RabbitMQ, RocketMQ), caching strategies (Redis, Memcached), distributed frameworks (Dubbo, Spring Cloud), and database optimization techniques (sharding, replication). The project serves as an interview preparation resource and learning material for Java engineers aiming to advance their skills in building large-scale, highly available systems.

Lazygit is a simple terminal UI for git commands that helps developers manage their repositories more efficiently through an interactive interface rather than typing git commands. Built in Go, it provides a user-friendly way to perform common git operations like staging, committing, branching, and pushing changes directly from the terminal. With nearly 79,000 stars, it's a widely-used tool designed to streamline git workflows for developers who prefer keyboard-driven interfaces and visual feedback.

LobeHub is an open-source AI agent platform that organizes and manages multiple AI agents as a unified team, enabling hiring, scheduling, and collaboration without requiring constant user oversight. Built with TypeScript and Next.js, it provides a comprehensive framework for creating, deploying, and orchestrating AI agents with support for multiple AI models, 10,000+ integrable skills via MCP plugins, and seamless multi-agent collaboration. The platform serves both individual users and professional developers who want to build intelligent agent teams that can work autonomously while remaining under human control. LobeHub emphasizes co-evolution between humans and agents, offering features like agent groups, shared workspaces, and sophisticated scheduling capabilities.

code-server is a tool that runs Visual Studio Code in a web browser, allowing developers to code on any device with a consistent development environment from anywhere. It enables users to leverage cloud servers for intensive tasks like testing and compilation while preserving battery life on local machines, making it ideal for remote development workflows. The project supports multiple deployment options including Docker, cloud providers, and team management through the Coder platform.

The TensorFlow Model Garden is an official repository providing state-of-the-art (SOTA) model implementations and examples built with TensorFlow, organized into official, research, and community sections. It demonstrates best practices for modeling across computer vision and natural language processing domains, with officially maintained implementations using TensorFlow 2's high-level APIs, research implementations by contributors, and integration with tools like TensorBoard for transparency and reproducibility. The repository includes the Orbit library for flexible training loops and is designed to help TensorFlow users leverage the framework effectively for both research and production development.

Last30days is an AI agent skill that aggregates real-time information from multiple social platforms (Reddit, X, YouTube, TikTok, Hacker News, Polymarket, and more) to synthesize comprehensive research briefs on any topic. Rather than relying on traditional search engines or editorial curation, it scores and ranks information by actual engagement metrics—upvotes, likes, transcripts, and real-money prediction market odds—giving users what people are genuinely talking about in the last 30 days. Designed for AI coding assistants like Claude Code, it works zero-config out of the box and supports 50+ agent skill hosts, making it useful for pre-meeting research, competitive analysis, trend understanding, and real-time intelligence gathering.

Ventoy is an open-source bootable USB creation tool that allows users to copy ISO, WIM, IMG, VHD, and EFI files directly to a USB drive without formatting, supporting over 1,300 tested operating systems across Windows, Linux, Unix, ChromeOS, and more. It provides a unified boot menu for selecting among multiple images and supports advanced features like secure boot, persistence, auto-installation, and multiboot across x86 Legacy BIOS, UEFI (IA32/x86_64/ARM64/MIPS64EL), and both MBR and GPT partition styles. The tool is designed for IT professionals, system administrators, and power users who need a flexible, non-destructive way to manage bootable media.

Elasticsearch is a free and open-source distributed search and analytics engine built in Java, optimized for speed and relevance on production-scale workloads. It serves as the foundation for the Elastic Stack and enables use cases including full-text search, vector search, retrieval-augmented generation (RAG), log analysis, metrics, and application performance monitoring. The repository contains the complete source code for building and deploying Elasticsearch, with comprehensive testing infrastructure and documentation for both local development and production deployments.

Awesome Public Datasets is a curated, topic-centric collection of high-quality open datasets covering fields like agriculture, biology, economics, and more. It serves researchers, data scientists, and analysts looking for reliable public data sources for analysis and machine learning projects. The repository is community-driven, automatically generated, and includes thousands of carefully vetted datasets organized by category with metadata and accessibility information.

NestJS is a progressive Node.js framework for building efficient, scalable, and enterprise-grade server-side applications using TypeScript. It combines Object-Oriented Programming, Functional Programming, and Functional Reactive Programming paradigms, while providing compatibility with Express and Fastify. Designed for developers who need a structured, testable architecture out of the box, NestJS is ideal for building microservices, REST APIs, and complex backend systems with strong organizational patterns inspired by Angular.

Hello-Agents is a comprehensive, free open-source tutorial by Datawhale that teaches how to build AI-native intelligent agents from first principles to practical implementation. The course covers foundational theory, classic agent paradigms (ReAct, Plan-and-Solve, Reflection), low-code platforms (Dify, Coze, n8n), popular frameworks (AutoGen, AgentScope, LangGraph), and advanced topics like memory systems, RAG, context engineering, and agentic reinforcement learning. It combines theoretical understanding with hands-on practice through real-world projects like intelligent travel assistants and cyber towns, designed to transform learners from LLM users into agent system builders.

OpenHands is an AI-driven development platform that enables developers to build, run, and scale autonomous software agents powered by large language models like Claude and GPT. It offers multiple interfaces including a Python SDK for composable agent development, a CLI for command-line usage, a local GUI, and a cloud-hosted version with enterprise features like Slack/Jira integrations and multi-user support. The platform is designed for individual developers and enterprises seeking to automate software development tasks, achieving 77.6% on the SWEBench benchmark and trusted by major companies including TikTok, Netflix, and Amazon.

Redis is an open-source, in-memory data structure server that serves as a high-performance cache, NoSQL database, message broker, and vector search engine for real-time applications. Written in C, it provides sub-millisecond latency and supports diverse data types including strings, hashes, sets, sorted sets, JSON, streams, and vectors, making it ideal for caching, session management, queuing, real-time analytics, and AI applications. Developers choose Redis for its exceptional speed, flexibility, extensive command set, and proven scalability across millions of production deployments worldwide.

Grafana is an open-source observability and data visualization platform that enables users to query, visualize, alert on, and understand metrics from multiple data sources like Prometheus, Elasticsearch, InfluxDB, and PostgreSQL. It provides powerful dashboarding capabilities, dynamic alerts, log exploration, and multi-source data integration, making it essential for teams practicing data-driven monitoring and observability. Built with TypeScript and Go, Grafana serves as a comprehensive solution for enterprises and developers to visualize and understand their infrastructure and application metrics.

Union is a trust-minimized, zero-knowledge bridging protocol that enables censorship-resistant cross-chain communication and asset transfers across multiple ecosystems including Ethereum, Cosmos, and other EVM chains. Built with Rust and Go, it implements consensus verification without relying on trusted third parties, oracles, or multi-signatures, making it suitable for decentralized finance applications that require high security. The repository contains a comprehensive modular infrastructure including the Union node (uniond), zero-knowledge prover (galoisd), relayer (voyager), smart contracts, and web interfaces, all managed through reproducible Nix builds.

The OpenAI Cookbook is a comprehensive collection of example code and practical guides for using the OpenAI API, including GPT-4, ChatGPT, DALL-E, Whisper, and embeddings. Written primarily in Python through Jupyter notebooks, it provides developers with ready-to-use examples covering common tasks like function calling, retrieval-augmented generation (RAG), fine-tuning, agents, and prompt engineering. The repository serves as an educational resource for both beginners and advanced users looking to integrate OpenAI's models into their applications.

Dear ImGui is a bloat-free C++ library for creating immediate-mode graphical user interfaces designed for developers building tools, editors, and visualization applications rather than end-user software. It minimizes state management complexity, integrates seamlessly into 3D graphics pipelines across multiple rendering backends (OpenGL, Vulkan, DirectX, Metal, etc.), and is widely used by game studios and real-time applications for both in-game tooling and standalone applications. The library prioritizes fast iteration, ease of use, and portability while maintaining high performance with minimal dependencies.

Awesome Vue is a comprehensive, community-curated collection of high-quality resources, libraries, components, and tools for Vue.js developers. It serves as a centralized directory covering everything from official documentation and tutorials to UI component libraries, state management solutions, dev tools, and real-world projects built with Vue. With over 73,000 stars, this repository is an essential reference guide for both beginners learning Vue.js and experienced developers seeking quality packages and best practices.

Apache Superset is a modern, enterprise-ready business intelligence and data visualization platform that enables users to explore data, build interactive charts, and create dashboards without coding. It supports connections to nearly any SQL database or data engine and provides both a no-code interface and a powerful SQL editor for advanced querying. Superset is designed for teams looking to replace or augment proprietary BI tools with an open-source, cloud-native, and highly extensible solution.

This is the official CPython repository containing the source code for Python 3.16.0 alpha, the reference implementation of the Python programming language. It serves as the central hub for Python development and is maintained by the Python Software Foundation, providing the core interpreter, standard library, and comprehensive documentation. The repository welcomes contributions from developers worldwide and includes extensive build instructions, testing frameworks, and CI/CD pipelines for cross-platform support on Unix, Linux, macOS, Windows, and other systems.

Caddy is a fast, extensible HTTP/1-2-3 web server written in Go that prioritizes automatic HTTPS and ease of use. It features automatic certificate management via Let's Encrypt and ZeroSSL, a simple configuration syntax (Caddyfile), and a powerful JSON API for dynamic configuration. Designed for production use with zero external dependencies, Caddy scales to manage millions of TLS certificates and serves as both a standalone web server and reverse proxy.