The AI repos worth starring
Claude & MCP
Official example skills (SKILL.md files) you can read, copy and adapt for your own Claude workflows.
“Public repository for Agent Skills”
Anthropic's agentic coding tool for the terminal; the home of issues, releases and plugin examples.
“Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.”
A big community index of MCP servers, sorted by category.
“A collection of MCP servers.”
Reference MCP servers for files, Git, search and more: the quickest way to give an AI assistant real tools.
“Model Context Protocol Servers”
Runnable notebooks for tool use, RAG, vision and agents with the Claude API.
“A collection of notebooks/recipes showcasing some fun and effective ways of using Claude.”
Agents & frameworks
The most widely used toolkit for wiring LLMs to data, tools and prompts.
“The agent engineering platform.”
Lets an AI agent drive a real web browser: click, type, scroll and read pages.
“Open-source Python library for AI browser agents, plus Browser Use Cloud: stealth browsers and a hosted agent API.”
Microsoft's framework for multi-agent conversations and workflows.
“A programming framework for agentic AI”
Teams of role-playing agents that split up a task and work through it together.
“Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.”
Build agents as graphs with state, memory and human-in-the-loop steps.
“Build resilient agents.”
OpenAI's lightweight SDK for agents with tools, handoffs and guardrails.
“A lightweight, powerful framework for multi-agent workflows”
Run models locally
The easiest way to download and run open models on your own laptop.
“Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.”
The standard library for loading and fine-tuning open models.
“🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training. ”
A polished ChatGPT-style interface for local and hosted models.
“User-friendly AI Interface (Supports Ollama, OpenAI API, ...)”
Fast LLM inference in plain C/C++; the engine under many local AI apps.
“LLM inference in C/C++”
High-throughput serving for open models in production.
“A high-throughput and memory-efficient inference and serving engine for LLMs”
Prompting & learning
A huge community collection of role prompts for ChatGPT and other chat models.
“f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source, self-host for your organization with complete privacy.”
A free course on building generative AI apps, lesson by lesson.
“21 Lessons, Get Started Building with Generative AI ”
Build a GPT-style model step by step to really understand how LLMs work.
“Implement a ChatGPT-like LLM in PyTorch from scratch, step by step”
A thorough, free guide to prompt engineering techniques with papers and examples.
“🐙 Guides, papers, lessons, notebooks and resources for prompt engineering, context engineering, RAG, and AI Agents.”
Practical examples and guides for building with OpenAI's models.
“Examples and guides for using the OpenAI API”
The simplest, fastest repo for training a small GPT; great for learning the internals.
“The simplest, fastest repository for training/finetuning medium-sized GPTs.”
Apps & automation
Open-source workflow automation with native AI nodes; self-host it or use the cloud.
“Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.”
Build and run LLM apps, agents and RAG pipelines from a visual editor.
“Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.”
An open-source engine for chat-with-your-documents (RAG) with deep document parsing.
“RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs”
A modern, self-hostable chat app that works with many model providers.
“🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.”
Drag-and-drop builder for AI agents and chatbots.
“Build AI Agents, Visually”
Image & video
The classic web UI for Stable Diffusion with a huge extension ecosystem.
“Stable Diffusion web UI”
Node-based workflows for image and video generation; the power user's choice.
“The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.”
Midjourney-style image generation that runs locally with minimal setup.
“Focus on prompting and generating”
State-of-the-art diffusion models for images, video and audio in one library.
“🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.”
People to follow
Clear, from-first-principles code and courses on how neural networks and LLMs really work.
Prolific builder of practical LLM tools (like the `llm` CLI) and a must-read blog on AI.
Author of Build a Large Language Model (From Scratch); deep, readable ML teaching.
Implements new AI research papers in clean PyTorch, often within days.
Creator of llama.cpp and whisper.cpp, which made local AI fast on ordinary hardware.
Co-founder of LangChain; follow for agent tooling and patterns.
Co-founder of LlamaIndex; RAG and document agents.
Creator of Instructor for structured LLM outputs; practical RAG advice.