🌟 Leonxlnx/taste-skill caught my eye on GitHub Trending today.

🔗 https://github.com/Leonxlnx/taste-skill
📝 Taste-Skill - gives your AI good taste. stops the AI from generating boring, generic slop
──────────────────────────────

The Taste Skill repository on GitHub offers a collection of anti-slop agent skills designed to elevate the quality of AI-built interfaces. These skills focus on enhancing layout, typography, motion, and spacing to create more visually appealing and premium user interfaces. The repository includes a range of skills, such as design-taste-frontend, gpt-taste, and image-to-code-skill, each serving a specific purpose in the design and development process.

To use these skills, developers can install them via the npx skills add command, specifying the skill they wish to install. For example, to install the design-taste-frontend skill, you would run

npx skills add https://github.com/Leonxlnx/taste-skill --skill "design-taste-frontend"

. The repository also includes image generation skills that produce design images, which can be used in conjunction with ChatGPT Images or other agents that generate images.

The skills are highly customizable, with adjustable dials for design variance, motion intensity, and visual density, allowing developers to fine-tune the output to suit their needs. The Taste Skill project is backed by a community of sponsors and contributors, and its development is guided by research-focused writing found in the research/ directory.

Whether you're working on a new project or looking to improve an existing one, Taste Skill provides a powerful toolset to help you create more engaging, polished, and professional interfaces. So why settle for boilerplate UIs when you can taste the difference with Taste Skill?

──────────────────────────────
🧠 Channel: https://t.me/GithubRe

💡 cursor/plugins just hit the trending charts — here's why it matters.

🔗 https://github.com/cursor/plugins
📝 Cursor plugin specification and official plugins
──────────────────────────────

The cursor/plugins repository is a collection of official plugins for popular developer tools, frameworks, and SaaS products, each with its own .cursor-plugin/plugin.json manifest. These plugins offer a wide range of features, including incremental learning, team workflows, security audits, and code quality checks.

Key features of the repository include the continual-learning plugin for incremental transcript-driven memory updates, the cursor-team-kit for internal team workflows, and the thermos plugin for deep security and correctness audits.

To use these plugins, simply navigate to the desired plugin's directory and follow the instructions in the README.md file. The repository is structured as a multi-plugin marketplace, with each plugin having its own manifest and directory.

The target audience for this repository includes developers, DevOps teams, and anyone looking to extend the functionality of their developer tools.

From a technical standpoint, the plugins are built using a variety of technologies, including TypeScript and the @cursor/sdk library.

In short, the cursor/plugins repository is a powerful resource for anyone looking to streamline their development workflow and improve their code quality.
You can supercharge your development workflow with these plugins!

──────────────────────────────
🧠 Channel: https://t.me/GithubRe

🔥 run-llama/liteparse is trending — and it deserves your attention.

🔗 https://github.com/run-llama/liteparse
📝 A fast, helpful, and open-source document parser
──────────────────────────────

LiteParse is a fast and lightweight PDF parsing tool that provides high-quality spatial text parsing with bounding boxes. It's designed to run locally on your machine, without proprietary LLM features or cloud dependencies. Key features include fast text parsing, a flexible OCR system with built-in Tesseract and support for HTTP OCR servers, screenshot generation, and multiple output formats including JSON and text.

The tool is multi-language, allowing use from Rust, Node.js/TypeScript, Python, or the browser (WASM), and is multi-platform, supporting Linux, macOS (Intel/ARM), and Windows.

For installation, you can use your preferred package manager, with versions available for Node.js/TypeScript, Python, Rust, and the browser (WASM).

To use LiteParse, you can utilize the command-line interface (CLI) for parsing files, batch parsing, and generating screenshots.

LiteParse is perfect for those looking for a fast, local, and flexible PDF parsing solution - parse your documents like a pro, without the cloud.

──────────────────────────────
🧠 Channel: https://t.me/GithubRe

🔥 galilai-group/stable-worldmodel is trending — and it deserves your attention.

🔗 https://github.com/galilai-group/stable-worldmodel
📝 A platform for reproducible world model research and evaluation
──────────────────────────────

The stable-worldmodel GitHub repository provides a unified platform for world model research and evaluation. It offers a single interface for data collection, training, and evaluation with model-predictive control across various environments. The library ships with reference implementations of common baselines and planning solvers, allowing research code to focus on the model and objective.

Key features include:
- Unified interface for the three stages of world model research
- Reference implementations of common baselines and planning solvers
- Support for multiple data formats, including lance, hdf5, folder, video, and lerobot
- Environment suite with factors of variation for evaluating zero-shot generalization

To get started, you can install the library using pip install stable-worldmodel and follow the Quick Start guide. The library is suitable for researchers and engineers working on world model research and evaluation.

Technical highlights include:
- Modular design for easy extension and customization
- Support for multiple environments, including DeepMind Control Suite, Gymnasium classic control, and more
- Built-in tools for data collection, training, and evaluation

The target audience for this library includes:
- Researchers working on world model research and evaluation
- Engineers implementing and deploying world models in real-world applications

In summary, stable-worldmodel provides a powerful and flexible platform for world model research and evaluation, making it an ideal choice for researchers and engineers in the field. Streamline your world model research with stable-worldmodel.

──────────────────────────────
🧠 Channel: https://t.me/GithubRe

📌 Spotted on GitHub Trending: byoungd/English-level-up-tips — let's break it down.

🔗 https://github.com/byoungd/English-level-up-tips
📝 An advanced guide to learn English which might benefit you a lot 🎉 . 离谱的英语学习指南/英语学习教程/英语学习/学英语
──────────────────────────────

The English-level-up-tips GitHub repository is a comprehensive guide to learning English, created by byoungd. This project aims to provide a detailed and structured approach to improving English skills, with a focus on efficiency and effectiveness. The guide covers various aspects of English learning, including vocabulary, listening, reading, speaking, and writing.

The repository also features a new AI chapter, which has been updated to 2026, providing a systematic approach to using AI tools like Gemini, ChatGPT, and Claude for English learning. The guide is designed to help learners discover the joy of learning English and make progress in a natural and enjoyable way.

The target audience for this repository includes English language learners of all levels, from beginners to advanced learners. The guide is suitable for anyone looking to improve their English skills, whether for personal or professional purposes.

The key technical highlights of the repository include its comprehensive coverage of English learning topics, its use of AI tools, and its structured approach to learning. The repository also features a range of resources, including word lists, learning tips, and personal stories.

In conclusion, the English-level-up-tips repository is a valuable resource for anyone looking to improve their English skills. With its comprehensive guide, AI-powered tools, and personal approach, it has the potential to make a significant impact on English language learning. So, why not start your English learning journey today and make it a fun and rewarding experience? Learning is, after all, one of life's greatest joys!

──────────────────────────────
🧠 Channel: https://t.me/GithubRe

🌟 Biohub/esm caught my eye on GitHub Trending today.

🔗 https://github.com/Biohub/esm
📝 No description.
──────────────────────────────

The Biohub/esm GitHub repository is home to a groundbreaking project that's changing the face of protein biology. At its core, it's a world model for protein biology, comprising three main components: ESMC, ESMFold2, and ESM Atlas.

ESMC is a state-of-the-art protein language model that's been trained on billions of protein sequences, learning the rules of protein biology. It's available through the Biohub Platform or Hugging Face, allowing users to run inference with minimal setup.

ESMFold2 is a state-of-the-art structure prediction model that combines ESMC embeddings with a diffusion-based architecture, predicting high-resolution protein structures directly from amino acid sequences.

ESM Atlas is a map of 6.8 billion proteins, organized according to the internal world model of ESMC. It's made interpretable through sparse autoencoders, revealing functional relationships between proteins.

These tools are designed for researchers, scientists, and developers working in the field of protein biology, providing a powerful platform for prediction, design, and discovery.

One-liner takeaway: The Biohub/esm repository is revolutionizing protein biology with its cutting-edge models and tools, empowering scientists to unlock new discoveries and advancements in the field.

──────────────────────────────
🧠 Channel: https://t.me/GithubRe

🔍 Deep-diving into Crosstalk-Solutions/project-nomad — fresh off the trending list.

🔗 https://github.com/Crosstalk-Solutions/project-nomad
📝 Project N.O.M.A.D, is a self-contained, offline survival computer packed with critical tools, knowledge, and AI to keep you informed and empowered—anytime, anywhere.
──────────────────────────────

Project N.O.M.A.D. is a self-contained, offline-first knowledge and education server that keeps you informed and empowered—anytime, anywhere. It's packed with critical tools, knowledge, and AI, including a local AI chat, offline Wikipedia, medical references, ebooks, education platform, offline maps, data tools, and notes.

Key features include:
- AI chat with knowledge base
- Information library with offline Wikipedia and medical references
- Education platform with Khan Academy courses
- Offline maps and data tools
- Notes and system benchmark

Usage is straightforward: just install it on a Debian-based operating system (like Ubuntu), access it through your browser, and start exploring.

From a technical standpoint, Project N.O.M.A.D. is built using Docker and includes a management UI and API that orchestrates a collection of containerized tools and resources.

The target audience includes anyone looking for a reliable, offline knowledge and education server, from students to researchers and outdoor enthusiasts.

In a nutshell: Project N.O.M.A.D. is your ultimate offline knowledge companion - take the power of knowledge with you, anywhere.

──────────────────────────────
🧠 Channel: https://t.me/GithubRe