What is awesome-agentic-ai-zh?

This project is a curated learning roadmap for understanding AI agents, written in Traditional Chinese, Simplified Chinese, and English. It is not code you run, but a structured guide that collects over 145 curated projects, resources, and short hands-on exercises, meant to take a reader from knowing nothing about large language models all the way to being able to design systems where multiple AI agents work together. The roadmap is organized into 8 stages. The first three stages, covering basics like Python, working with APIs, and prompt design, are shared by everyone. After that, learners choose one of two paths. The first path is for people who mainly want to use existing command line AI agent tools, such as Claude Code or similar tools, more effectively in their daily work. The second path is for people who want to build their own agents from scratch, covering topics like agent frameworks, tool use, giving an agent memory through retrieval, and coordinating multiple agents together. Both paths eventually meet again at shared stages covering the Claude Code ecosystem and different ways an agent can interact with a computer, such as browsing the web or running code in a sandbox. Each stage links out to curated external projects and includes small practice exercises, generally 70 to 150 lines of example code, along with side by side comparisons using either local AI models or provider APIs like Anthropic's. Beyond the two main learning paths, the project also includes shorter, role specific reading lists aimed at researchers, software developers, teachers, knowledge workers, and everyday users who simply want to use AI tools like ChatGPT without necessarily writing code. The project estimates the CLI focused path takes about 8 to 10 weeks, while the path for building agents from scratch takes roughly 16 to 22 weeks of core material, or about 5 to 7 months at a part time pace. This project is released under the MIT license and all its content is offered for free.