Github Top Repositories

⚡ Comfy-Org/ComfyUI is making waves. Here's the full picture. 🔗 https://github.com/Comfy-Org/ComfyUI 📝 The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. ────────────────────────────── ComfyUI is the most powerful and modular AI engine for content creation, offering a node graph interface for generating images, videos, 3D models, audio, and more. It natively supports state-of-the-art models and provides API nodes for closed source models. Available on Windows, Linux, and macOS, ComfyUI can be used locally with a desktop application or portable install, or on the Comfy Cloud. Its key features include a visual node graph, reusable subgraphs, and efficient local execution. ComfyUI is suitable for visual professionals who demand control over every aspect of their work. With its broad native model support and customizable nodes, ComfyUI is a versatile tool for content creation. Try it out and experience the power of AI-driven content creation - ComfyUI: where creativity meets innovation. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

📌 Spotted on GitHub Trending: google/skills — let's break it down. 🔗 https://github.com/google/skills 📝 Agent Skills for Google products and technologies ────────────────────────────── The google/skills GitHub repository is a collection of Agent Skills for various Google products and technologies, including Google Cloud. The repository is under active development and provides a wide range of skills, from getting started with Google Cloud to more advanced topics like AI/ML, infrastructure, and management tools. To use these skills, you can install them using the command npx skills add google/skills, and then select the specific skills you want to install. The repository also includes plugins for agent harnesses like Claude Code, Codex, and Antigravity CLI. The skills are organized into categories, including AI/ML, Infrastructure, Databases and analytics, and Security and identity, among others. Each skill provides a specific set of instructions or guidelines for completing a task or achieving a particular outcome. The repository is maintained by Google and is open to contributions from the community. If you need help or encounter issues with these skills, you can search for existing issues or open a new one in the GitHub Issue Tracker. Overall, the google/skills repository is a valuable resource for anyone looking to develop skills and expertise in Google products and technologies. With its wide range of skills and plugins, it's a great way to get started with Google Cloud and other Google technologies. You can npx skills add google/skills and start building your skills today! ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🚀 Meet goauthentik/authentik: a gem from today's GitHub trending list. 🔗 https://github.com/goauthentik/authentik 📝 The authentication glue you need. ────────────────────────────── authentik is an open-source Identity Provider (IdP) for modern Single Sign-On (SSO) solutions. It supports various protocols like SAML, OAuth2/OIDC, LDAP, and RADIUS, making it suitable for self-hosting in small labs to large production clusters. The project offers different installation methods, including Docker Compose for small setups and Kubernetes (Helm Chart) for larger ones. Technical highlights include a MIT License for the open-source version and an enterprise offering for organizations. For developers, the project provides a Developer Documentation section with guidelines for setting up local build environments and testing contributions. authentik is perfect for organizations looking for a robust, large-scale identity management solution. One-liner takeaway: Replace existing IdPs with authentik for a robust, open-source SSO solution. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🔥 ZhuLinsen/daily_stock_analysis is trending — and it deserves your attention. 🔗 https://github.com/ZhuLinsen/daily_stock_analysis 📝 LLM 驱动的多市场股票智能分析系统：多源行情、实时新闻、决策看板与自动推送，支持零成本定时运行。 LLM-powered multi-market stock analysis system with multi-source market data, real-time news, decision dashboard, automated notifications, and cost-free scheduled runs. ────────────────────────────── Daily Stock Analysis is a powerful AI-driven system for analyzing stocks from various markets, including A-share, Hong Kong, US, Japan, Korea, and Taiwan. This open-source project provides a web-based interface for configuring and monitoring analysis tasks, as well as a command-line interface for manual execution. The system offers key features such as: - AI decision reports with core conclusions, scores, trends, and risk alerts - Multi-market data aggregation, including quotes, K-lines, technical indicators, news, and basic face data - Web and desktop workspaces for manual analysis, task progress, historical reports, and configuration management - Agent strategy inquiry with 15 built-in strategies, supporting web, bot, and API interactions To use the system, simply fork the repository, configure your secrets, and enable GitHub Actions. You can also run the system locally or deploy it using Docker. The project has a strong technical stack, including AI models from Anspire, AIHubMix, and OpenAI, as well as data sources like TickFlow, AkShare, and Tushare. Main audience for this project includes: - Investors and traders seeking AI-driven insights - Researchers and developers interested in natural language processing and machine learning - Financial analysts looking for automated tools to support their work Technical highlights of the project include: - Support for multiple AI models and data sources - Flexible configuration options for customization - Automated reporting and notification capabilities Check out the project's documentation for more information on getting started, configuration options, and advanced features. In summary, Daily Stock Analysis is a robust and customizable platform for analyzing stocks using AI-driven insights. With its flexible configuration options, automated reporting, and support for multiple AI models and data sources, it's an ideal tool for investors, researchers, and financial analysts. Get started today and take your investment decisions to the next level! ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🌟 addyosmani/agent-skills caught my eye on GitHub Trending today. 🔗 https://github.com/addyosmani/agent-skills 📝 Production-grade engineering skills for AI coding agents. ────────────────────────────── Agent Skills is a GitHub repository that provides production-grade engineering skills for AI coding agents. The skills encode workflows, quality gates, and best practices that senior engineers use when building software, ensuring that AI agents follow them consistently across every phase of development. Key features include 24 skills that cover the development lifecycle, from defining what to build to shipping to production. These skills are packaged as slash commands, allowing for easy activation and automation of tasks. To get started, you can install the skills using the open skills CLI or integrate them with various agents such as Claude Code, Cursor, Codex, and more. From a technical standpoint, the skills are designed to be flexible and adaptable to different development environments and workflows. They are written in Markdown, making them easy to read and understand. The repository is suitable for developers and engineering teams looking to improve their development workflows and ensure consistency and quality in their code. In summary, Agent Skills is a valuable resource for anyone looking to streamline their development process and improve the quality of their code. With its flexible and adaptable design, it's a great tool for teams and individual developers alike. Automate your development workflow with Agent Skills and take your coding to the next level! ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

⚡ google-deepmind/weathernext is making waves. Here's the full picture. 🔗 https://github.com/google-deepmind/weathernext 📝 No description. ────────────────────────────── The google-deepmind/weathernext GitHub repository is home to the WeatherNext family of models, including WeatherNext 2 (WN2), a global, medium-range atmospheric and cyclone forecasting model developed by Google DeepMind and Google Research. The repo contains code for WN2, as well as prior generation models like GraphCast and GenCast. Key features include access to daily data feeds of WN2 model outputs across multiple platforms, such as Google Cloud, WeatherLab, and OpenMeteo. To get started, users can run the interactive Colab Notebook, which defaults to WeatherNext Cyclones Mini, or run the other models enumerated in the repository. The repository also provides pre-trained model weights and sample data on the Google Cloud Bucket. Technical highlights include the use of graph neural networks, diffusion-based ensemble forecasting, and autoregressive rollout steps to generate forecast predictions. The repository is suitable for researchers and developers interested in experimenting with the published models. However, please note that this is research code provided as-is, and there are no guarantees of API stability. If you use WeatherNext 2 in your research, please cite the relevant papers. One-liner takeaway: Dive into the google-deepmind/weathernext repository to explore the latest advancements in weather forecasting models, including WeatherNext 2, and discover how to run and experiment with these models. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🔍 Deep-diving into pranshuparmar/witr — fresh off the trending list. 🔗 https://github.com/pranshuparmar/witr 📝 Why is this running? Trace any process, port, container, or file back to what started it - CLI + TUI. ────────────────────────────── The witr tool is designed to answer a simple yet crucial question: Why is this running? It helps you trace any process, port, container, or file back to its origin, providing a clear understanding of what started it and what chain of systems is responsible for its existence. Key features include: - machine-readable JSON output - An interactive TUI dashboard for easy navigation - Support for various operating systems, including Linux, macOS, Windows, and FreeBSD To get started, you can install witr using your preferred package manager, such as Homebrew, Conda, or Winget, or by running the quick install script. Technical highlights include: -

curl -fsSL https://raw.githubusercontent.com/pranshuparmar/witr/main/install.sh | bash

for Unix-based systems -

irm https://raw.githubusercontent.com/pranshuparmar/witr/main/install.ps1 | iex

for Windows witr is perfect for developers, system administrators, and curious users who want to understand what's running on their systems and why. In a nutshell, witr is the missing link between what's running and why it's running, and it's a game-changer for anyone looking to take control of their system processes! ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🎯 msitarzewski/agency-agents landed on trending. Worth a proper look. 🔗 https://github.com/msitarzewski/agency-agents 📝 A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables. ────────────────────────────── The Agency is a collection of specialized AI agents that can transform your workflow. Each agent has a unique personality, expertise, and deliverables. You can browse the roster and install agents into various tools like Claude Code, Cursor, and Codex using the Agency Agents app or scripts. The app is available for macOS, Linux, and Windows and allows you to install and update agents with a click. The agents are production-ready and can be used for various tasks such as frontend development, backend architecture, and AI engineering. The Agency Agents repo provides a quick start guide and usage examples to get you started. Whether you're a developer, engineer, or entrepreneur, The Agency can help you streamline your workflow and improve productivity. Take your workflow to the next level with The Agency - your dream team of AI specialists. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🎯 vitali87/code-graph-rag landed on trending. Worth a proper look. 🔗 https://github.com/vitali87/code-graph-rag 📝 The ultimate RAG for your monorepo. Query, understand, and edit multi-language codebases with the power of AI and knowledge graphs ────────────────────────────── Code-Graph-RAG is a powerful tool that parses a multi-language codebase, builds a knowledge graph of its structure, and allows you to query, edit, and optimize that code in plain English. With support for languages like Python, TypeScript, and Java, it works across a monorepo of mixed languages under one unified graph schema. Key features include natural language querying, AST-based surgical patching, and optimization against language best practices. The system has two components: a multi-language parser and a RAG system that turns natural language into Cypher queries. To get started, you can install Code-Graph-RAG using pipx install "code-graph-rag[treesitter-full,semantic]" and then parse a repository into the graph using cgr start --repo-path /path/to/repo --update-graph. The project is well-documented, with guides for getting started, user guide, architecture, and advanced topics. Code-Graph-RAG is open source and free to use, with enterprise services available for organizations that need more. In short, Code-Graph-RAG revolutionizes code analysis and optimization - parse, query, and edit your codebase with ease. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

📌 Spotted on GitHub Trending: PrimeIntellect-ai/prime-agent — let's break it down. 🔗 https://github.com/PrimeIntellect-ai/prime-agent 📝 A self-improving RLM agent for coding workflows and long-running autonomous tasks. ────────────────────────────── Introducing Prime Agent, a self-improving Recursive Language Model (RLM) agent designed for long-running work and research. This open-source tool combines a persistent Python environment with a durable harness state, allowing useful context and operating patterns to outlive a single session. Key Features: - Recursive Language Model (RLM) treats context as variables and tools like recursive subagents as function calls - Continual Harness stores supplemental prompts, memories, and reusable subagent specifications as durable state - Programmatic tooling with persistent IPython and subagents for parallel or background work - Autonomous mode with configurable turn, token, and time budgets - Direct agent-to-agent communication for orchestrating work Usage: Install Prime Agent using the provided install.sh script, then run it in your project directory. Use the prime-agent command to manage sessions, attach to running agents, and inspect background services. Technical Highlights: - Daemon-backed continuity keeps sessions running when the terminal detaches - Heartbeats and schedules for re-entering sessions periodically or at a specific time - Persistent goals keep objectives active across turns until completion or pause Audience: Prime Agent is designed for researchers, developers, and anyone looking for a powerful tool for long-running work and automation. One-liner takeaway: Prime Agent revolutionizes long-running work by combining a persistent Python environment with a self-improving RLM agent, making it a game-changer for researchers and developers alike. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

💡 bannedbook/fanqiang just hit the trending charts — here's why it matters. 🔗 https://github.com/bannedbook/fanqiang 📝 翻墙-科学上网 ────────────────────────────── The bannedbook/fanqiang GitHub repository is a comprehensive resource for scientific internet access and wall-crossing tools. It provides a wide range of free and open-source software, including Chrome one-click wall-crossing packages, Android and iOS apps, and tutorials for various operating systems. The repository also offers free SS accounts and v2ray accounts, as well as guides for setting up and using these tools. Key features include: - Chrome one-click wall-crossing package with integrated tools like Goflyway, v2ray, and SSR - Android and iOS apps for wall-crossing, such as V2free and Shadowsocks - Tutorials for Windows, macOS, Linux, and game consoles - Free SS and v2ray accounts for users The target audience includes individuals seeking to access the internet freely and securely, particularly those in regions with restricted internet access. The takeaway: Break down the walls and access the internet freely with bannedbook/fanqiang! ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🚀 Meet litu54/DevOps-Interview-Guide: a gem from today's GitHub trending list. 🔗 https://github.com/litu54/DevOps-Interview-Guide 📝 DevOps Interview Guide ────────────────────────────── The litu54/DevOps-Interview-Guide GitHub repository is a treasure trove for DevOps and SRE professionals. It contains real interview questions from various companies, including 85 well-known organizations, collected from actual interview experiences in 2025 and 2026. The repository is meticulously organized, with each company having its own folder, and each interview experience stored in a separate file. You can browse through the repository to prep for your interview by searching for a specific company or skimming through multiple folders. The repository covers a wide range of topics, including Kubernetes, Docker, Terraform, AWS/Azure/GCP, CI/CD, Ansible, Linux, and SRE fundamentals. The repository is constantly updated with new interview experiences, and you can even contribute your own experience to help others. Whether you're a seasoned professional or just starting out, this repository is an invaluable resource for anyone looking to ace their DevOps or SRE interview. One-liner takeaway: Get ahead of the game with the litu54/DevOps-Interview-Guide, and land your dream DevOps or SRE job with confidence! ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

🔥 denoland/celld is trending — and it deserves your attention. 🔗 https://github.com/denoland/celld 📝 self-hosted, distributed Durable Objects ────────────────────────────── celld is an open-source daemon that runs Cloudflare Workers and Durable Objects on your own machines. It allows you to self-host and distribute Durable Objects, with each object being its own SQLite database addressed by name and replicated to an S3-compatible bucket. Key features of celld include: - Self-hosted: run your own Durable Objects - Distributed: replicate objects across multiple nodes - SQLite databases: each object is its own database - S3-compatible bucket: store and replicate object data To use celld, you can install it using the provided installer script, or run it from a Docker container. You can then deploy your applications to an S3-compatible bucket and start celld against that bucket. From a technical standpoint, celld uses the standard AWS credential chain and embeds V8 to execute Wrangler bundles. It also uses object-storage compare-and-swap to ensure that exactly one node owns a cell at a time, without a membership protocol or consensus service. celld is suitable for developers and organizations looking to self-host and distribute their Durable Objects, with a focus on scalability, reliability, and security. In summary: Take control of your Durable Objects with celld - the self-hosted, distributed solution for Cloudflare Workers and Durable Objects. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

📌 Spotted on GitHub Trending: LadybirdBrowser/ladybird — let's break it down. 🔗 https://github.com/LadybirdBrowser/ladybird 📝 Truly independent web browser ────────────────────────────── The Ladybird browser is a game-changer - a truly independent web browser built on a novel engine based on web standards. It's still in pre-alpha, so it's only suitable for developers right now. One of the key features of Ladybird is its multi-process architecture, which includes a main UI process, several WebContent renderer processes, an ImageDecoder process, and a RequestServer process. This means that image decoding and network connections are done out of process, making it more robust against malicious content. Each tab has its own renderer process, sandboxed from the rest of the system, adding an extra layer of security. Ladybird also inherits many core library support components from SerenityOS, including LibWeb, LibJS, and LibWasm. To get started with Ladybird, you can find build instructions in the Documentation folder, and the project runs on Linux, macOS, Windows (with WSL2), and many other *Nixes. If you're interested in participating in the project, you can join the Discord server to discuss issues and development, and read the getting started guide to learn how to contribute. Ladybird is licensed under a 2-clause BSD license. In short, Ladybird is a browser that's all about independence and security - and it's an exciting project to watch, or even contribute to! ────────────────────────────── 🧠 Channel: https://t.me/GithubRe

📌 Spotted on GitHub Trending: google/guava — let's break it down. 🔗 https://github.com/google/guava 📝 Google core libraries for Java ────────────────────────────── The Guava library is a set of core Java libraries from Google, featuring new collection types, immutable collections, a graph library, and utilities for concurrency, I/O, hashing, primitives, strings, and more. It comes in two flavors: JRE and Android. To use Guava, add a dependency to your project using Maven or Gradle. For example, with Maven:

<dependency>
  <groupId>com.google.guava</groupId>
  <artifactId>guava</artifactId>
  <version>33.6.0-jre</version>
</dependency>

Guava is suitable for Java developers looking to simplify their code and improve performance. Key technical highlights include its collection types, such as multimap and multiset, and utilities like ImmutableList. Guava's APIs are largely binary-compatible, ensuring backward compatibility. One-liner takeaway: Use Guava to simplify your Java development and boost performance with its robust collection types and utility classes. ────────────────────────────── 🧠 Channel: https://t.me/GithubRe