A structured, trilingual learning roadmap with over 240 curated resources that guides developers and general learners from basic LLM concepts to building multi-agent systems. It provides an 8-stage curriculum, practice exercises, and catalogs for MCP/Skills and subagent recipes.
Project overview
It offers a fully maintained trilingual documentation set (Traditional Chinese, English, Simplified Chinese) alongside a structured 8-stage learning path covering prompt engineering through multi-agent systems.
Project type
MCP · AI Agent · Learning Resources
Use cases
Learning & Education
Deployment
Refer to project documentation
License
MIT
Best for
Developers, general users, and researchers who want a structured, trilingual roadmap for progressing from LLM basics to multi-agent systems.
Learners who benefit from short, illustrative practice exercises comparing local orchestration via Ollama with remote inference via the Anthropic SDK.
Key capabilities
A learning roadmap structured into 8 stages that takes learners from basic LLM concepts and prompt engineering to building multi-agent systems.
Includes over 240 curated projects with details such as stars, target audience, teaching content, and execution instructions, alongside a catalog of 65+ MCP/Skills.
Simple illustrative practice exercises (70-150 lines) for each stage, featuring dual-path Ollama/Anthropic SDK comparisons and mock-based tests.
A directory of 65+ MCP servers and Skills, categorized into 16 major groups with star ratings.
Provides 15 copy-paste ready dispatch recipes for delegating tasks to subagents.
Limitations and risks
The exercises favor English terminology and assume basic Python knowledge, which may be difficult for absolute beginners who do not follow the Stage 0 foundations first.
Getting started
Setup difficulty is rated as easy. Users can clone the repository or read the hosted documentation online, starting by reading the README and the foundations stage.
Alternatives and comparisons
Provides an 18-lesson course with written guides, videos, and Python code samples to teach how to build AI Agents for beginners.
A 503-lesson open-source curriculum that builds AI engineering understanding from raw math to production agents, with every lesson producing a reusable artifact.
A comprehensive resource of 93+ production-ready projects and in-depth tutorials across all skill levels to experiment and succeed in AI engineering.