Github Top Repositories
14.1K subscribers
3.07K photos
59 videos
10 files
3.1K links
Top GitHub repositories in one place 🚀
Explore the best projects in programming, AI, data science, and more.
Download Telegram
Github Top Repositories
Photo
📌 Spotted on GitHub Trending: cathrynlavery/diagram-design — let's break it down.

🔗 https://github.com/cathrynlavery/diagram-design
📝 29 editorial diagram types for Claude Code. Self-contained HTML + SVG. No shadows, no Mermaid-slop.
──────────────────────────────

The Diagram Design GitHub repository offers a unique solution for creating high-quality, editorial-style diagrams that match your brand. This tool provides 27 visual types, including architecture, flowcharts, and timelines, all of which can be easily customized to fit your brand's colors and typography.

The repository includes a Claude Code skill that allows for seamless integration and can be installed using the plugin marketplace. The skill can also be used with Codex and Pi, making it versatile and accessible to various users.

One of the key features of Diagram Design is its ability to extract your brand's palette and font stack from your website, allowing for a consistent look and feel across all diagrams. The tool also performs automatic contrast checks to ensure that the diagrams are accessible and meet WCAG AA standards.

The target audience for this repository includes developers, designers, and anyone looking to create high-quality diagrams that match their brand's style. With its easy-to-use interface and customizable options, Diagram Design is an excellent choice for those who want to create professional-looking diagrams without spending hours designing them from scratch.

In summary, Diagram Design is a powerful tool that helps you create editorial-quality diagrams in minutes, with a consistent brand look and accessible design - and that's a game-changer for anyone who's ever struggled with creating diagrams that look great and communicate effectively!

──────────────────────────────
🧠 Channel: https://xn--r1a.website/GithubRe
1
📌 Spotted on GitHub Trending: cursor/plugins — let's break it down.

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

The Cursor plugins repository is a collection of official plugins for various developer tools, frameworks, and SaaS products. Each plugin has its own directory at the repository root, containing a .cursor-plugin/plugin.json manifest. The plugins offer a range of features, such as incremental transcript-driven memory updates, thermo-nuclear branch review, and rendering PR diffs as interactive canvases.

To use these plugins, you can simply navigate to the respective plugin directory and follow the instructions in the README.md file. The repository is structured as a multi-plugin marketplace, with a root .cursor-plugin/marketplace.json file that lists all available plugins.

The plugins are designed for developers and teams looking to streamline their workflows and improve productivity. With plugins for popular tools like Gmail, Google Drive, and Salesforce, as well as integrations with other services like HubSpot and Intercom, this repository has something for everyone.

Some of the key technical highlights include the use of MCP server definitions and agent skills to enable seamless integration with various services. The plugins are all licensed under the MIT license, making it easy to get started and contribute to the community.

Whether you're looking to build apps, scripts, or automations or simply want to connect to your favorite SaaS products, the Cursor plugins repository has got you covered. So why wait? Dive in and start exploring the world of Cursor plugins today - your productivity will thank you!

──────────────────────────────
🧠 Channel: https://xn--r1a.website/GithubRe
1
Github Top Repositories
Photo
cactus-compute/needle is making waves. Here's the full picture.

🔗 https://github.com/cactus-compute/needle
📝 14MB foundation model for tiny devices; phones, wearables, smart home, and robots.
──────────────────────────────

Needle 2 is an open, 45M-parameter model for tool calling and structured extraction, optimized for tiny devices. It's a single, self-contained 14MB binary that runs in about 28MB of RAM. Key features include a simple contract for tool calls, confidence-gated responses, and bounded memory usage.

To get started, you can install the Python package with pip install cactus-needle and describe your tools to integrate them with the model.
import needle
@needle.tool
def get_weather(city: str):
"Get the current weather for a city."
return {"city": city, "temp_c": 27, "sky": "clear"}


The model is technically notable for its use of a Simple Attention Network architecture, which includes a Hadamard MLP, GQA attention, and engram key-value memory.

The target audience for Needle 2 includes developers and researchers looking to integrate tool calling and structured extraction into their applications, particularly those with limited resources or memory constraints.

One-liner takeaway: Needle 2 packs a big punch in a tiny package, enabling powerful tool calling and structured extraction on even the smallest devices.

──────────────────────────────
🧠 Channel: https://xn--r1a.website/GithubRe
Github Top Repositories
Photo
🎯 unslothai/unsloth landed on trending. Worth a proper look.

🔗 https://github.com/unslothai/unsloth
📝 Local UI to run and train LLMs and diffusion models, including Qwen3.8, Kimi K3, MiniMax-H3, Gemma 4, DeepSeek-V4, FLUX and more.
──────────────────────────────

Unsloth is a cutting-edge desktop app that allows you to run and train AI models locally, with support for various model types. Its key features include running and training LLMs, diffusion, embedding, audio models, and more. You can use it to fine-tune models 2× faster with 70% less VRAM, and export or deploy models in different formats. Unsloth also offers remote access through secure Cloudflare HTTPS.

To get started, you can download the native Unsloth Desktop app for your operating system, or install it manually using curl or powershell commands. Unsloth is suitable for developers, researchers, and AI enthusiasts who want to work with AI models locally.

The app is available for Windows, macOS, and Linux, with support for multi-GPU setups and Vulkan acceleration. With Unsloth, you can unleash the full potential of AI models on your local machine - no cloud required!

──────────────────────────────
🧠 Channel: https://xn--r1a.website/GithubRe
📌 Spotted on GitHub Trending: public-apis/public-apis — let's break it down.

🔗 https://github.com/public-apis/public-apis
📝 A collective list of free APIs
──────────────────────────────

The public-apis/public-apis repository is a curated list of free APIs from various domains, maintained by the community and APILayer. It's a treasure trove of APIs that you can use for your own products. The repository includes APIs for animals, anime, anti-malware, and many more. You can browse the list of APIs, which includes information such as authentication requirements, HTTPS support, and CORS headers. To get started, you can fork the official APILayer Postman Collection and start building your project in under 60 seconds. The repository is open to contributions, and you can get involved by checking out the contributing guide. Join the community on Discord to get updates, ask questions, and get answers. With public-apis/public-apis, you can integrate production-grade REST APIs into your product using one account, one dashboard, and one API key. So, what are you waiting for? Dive into the world of public APIs and start building your next project today - one API at a time!

──────────────────────────────
🧠 Channel: https://xn--r1a.website/GithubRe
1
Github Top Repositories
Photo
💡 MakazhanAlpamys/Soup just hit the trending charts — here's why it matters.

🔗 https://github.com/MakazhanAlpamys/Soup
📝 Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.
──────────────────────────────

Soup is a game-changer for fine-tuning Large Language Models (LLMs). It simplifies the process into a single command, eliminating the need for SSH, complex configurations, and tedious infrastructure management. With Soup, you can fine-tune an 8B model on a 4 GB laptop GPU, thanks to its innovative layer streaming feature.

The key features of Soup include:
- Zero SSH: No more SSHing into broken GPU boxes.
- One config: A simple YAML file for all your needs.
- Auto everything: Batch size, GPU detection, and quantization are all handled automatically.
- Works locally: Train on your own GPU with QLoRA, no cloud required.

To get started with Soup, you can:
pip install "soup-cli[train]"
soup init --template chat
soup train


Soup is perfect for anyone looking to simplify their LLM fine-tuning workflow, from researchers to developers. Its ability to stream layers and quantify models makes it an ideal solution for those working with limited GPU resources.

In short, Soup is a powerful tool that makes fine-tuning LLMs a breeze, and its innovative features make it a must-try for anyone in the field. Train your LLMs with ease, and focus on what really matters – creating value for your users!

──────────────────────────────
🧠 Channel: https://xn--r1a.website/GithubRe
2
Github Top Repositories
Photo
💡 github/spec-kit just hit the trending charts — here's why it matters.

🔗 https://github.com/github/spec-kit
📝 💫 Toolkit to help you get started with Spec-Driven Development
──────────────────────────────

Spec Kit is an open source toolkit that enables spec-driven development for building high-quality software with any AI coding agent. It provides a ready-to-use, spec-driven process that's endlessly extensible, community-driven, and suitable for entire organizations.

The key features of Spec Kit include:
- Spec-Driven Development: Flips the script on traditional software development by making specifications executable.
- Specify CLI: A command-line interface for managing spec-driven development workflows.
- AI Coding Agent Integrations: Supports over 30 AI coding agents, including CLI tools and IDE-based assistants.
- Extensions and Presets: Allows for customization through extensions and presets, as well as project-local overrides.

To get started with Spec Kit, you can:
uv tool install specify-cli --from git+https://github.com/github/spec-kit.git@vX.Y.Z
Then, initialize a project with specify init my-project --integration copilot and establish project principles using /speckit.constitution.

Spec Kit is designed for developers, organizations, and communities looking to streamline their software development process with AI coding agents.

The technical highlights of Spec Kit include its ability to generate executable specifications, create technical implementation plans, and break down tasks for implementation.

In a nutshell, Spec Kit revolutionizes software development by making specifications the king, and code its servant - build what you specify, not the other way around.

──────────────────────────────
🧠 Channel: https://xn--r1a.website/GithubRe
Github Top Repositories
Photo
📌 Spotted on GitHub Trending: megadose/holehe — let's break it down.

🔗 https://github.com/megadose/holehe
📝 holehe allows you to check if the mail is used on different sites like twitter, instagram and will retrieve information on sites with the forgotten password function.
──────────────────────────────

Holehe is an OSINT tool that efficiently finds registered accounts from emails. It checks if an email is attached to an account on over 120 sites like Twitter, Instagram, and Imgur. The tool retrieves information using the forgotten password function without alerting the target email.

It runs on Python 3 and can be installed via PyPI, GitHub, or Docker. Holehe can be run from the CLI or embedded within existing Python applications.

For each module, data is returned in a standard dictionary format with details like rateLimit, exists, emailrecovery, and more. It's ideal for security researchers and developers.

The tool has a Maltego Transform and is built for educational purposes only under the GNU General Public License v3.0.

Get started with Holehe today and uncover hidden accounts - one email at a time!

──────────────────────────────
🧠 Channel: https://xn--r1a.website/GithubRe