SuperAPI AI Directory
Partner with us中文AI newsFavorites
All toolsAI codingawesome-code-docs

awesome-code-docs

awesome-code-docs is an in-depth tutorial collection for AI agents, LLM frameworks, programming tools, MCP, and open-source developer platforms, maintained by johnxie on GitHub. It includes 203 in-depth tutorials and emphasizes explaining how complex systems "work" and why they are designed that way, rather than only describing features. Each tutorial follows a unified 8-chapter structure, including Mermaid architecture diagrams, annotated code examples from real codebases, and quick-reference summary tables. The tutorials are organized into categories such as AI agents and multi-agent systems, LLM frameworks and RAG, LLM infrastructure and services, chat and AI applications, developer tools and productivity, database knowledge and analytics, systems and infrastructure, and voice and multimodal AI, and provide 11 learning paths covering topics from AI fundamentals, agent building, RAG engineering, full-stack AI, LLM infrastructure, AI programming tools, autonomous AI engineers, and MCP mastery to voice AI, the OpenAI ecosystem, and Vibe Coding platforms. The repository also provides navigation layers such as an A-Z directory, a query hub, an intent map, and market signals, and marks the source verification status of tutorials. The tutorials are generated by AI-driven codebase analysis and then enhanced through human review.

Open website

About this tool

Overview

awesome-code-docs is a GitHub tutorial collection maintained by johnxie, focused on AI agents, LLM frameworks, programming tools, MCP, and open-source developer platforms, with a total of 203 in-depth tutorials. Its purpose is not simply to list "how to do it," but to explain how complex systems operate underneath and why they are designed that way, emphasizing deep understanding through architecture diagrams, real code walkthroughs, and production-grade patterns.

Main Capabilities

  • Tutorial count and structure: It includes 203 in-depth tutorials, each following a unified 8-chapter structure; each chapter includes Mermaid architecture diagrams, annotated code examples from real codebases, and summary tables for quick reference.
  • Topic categories: The tutorial directory is grouped by topic, including AI agents and multi-agent systems, LLM frameworks and RAG, LLM infrastructure and services, chat and AI applications, developer tools and productivity, databases/knowledge and analytics, systems and infrastructure, voice and multimodal AI, and more.
  • Learning paths: It provides 11 learning paths, such as AI fundamentals, agent builder, RAG engineer, full-stack AI, LLM infrastructure, AI programming tools, autonomous AI engineer, MCP mastery, voice and speech AI, OpenAI ecosystem, and Vibe Coding platforms; some paths are marked with estimated duration and difficulty, such as the autonomous AI engineer path at 30-45 hours and advanced difficulty, the MCP mastery path at 100-135 hours and intermediate to advanced difficulty, the voice and speech AI path at 25-35 hours and intermediate difficulty, the OpenAI ecosystem path at 35-45 hours and beginner to intermediate difficulty, and the Vibe Coding platforms path at 35-50 hours and intermediate to advanced difficulty.
  • Navigation experience: To reduce context switching and dead ends, the tutorial index includes a "Navigation and Backlinks" section that links back to the main directory, A-Z directory, query hub, and category hubs, and fixes the Chapter 1 entry link so readers can directly enter the corresponding track. The repository also provides quick jump entries such as the Tutorials workspace guide, A-Z tutorial directory, and search intent map.
  • Source verification: All tutorial indexes were re-verified against upstream GitHub repositories on 2026-06-01; 203 tutorials were scanned, 203 tutorials had source repositories, 0 source repositories were unverified, and 214 unique verified source repositories were found. Verification artifacts include the tutorial source verification report.announcements, tutorial source verification JSON, and the verification script scripts/verify_tutorial_sources.py.
  • Ecosystem snapshot and trends: includes the Taskade ecosystem snapshot (verified on 2026-06-01), high-intent searches for Taskade/Genesis/AI/MCP; and the "Trending Vibe-Coding Repositories" (auto-updated on 2026-10-12), which displays real-time GitHub market signals for high-impact open-source coding agents and the Vibe-Coding ecosystem, with data sourced from the GitHub REST API (stargazers_count, pushed_at), refreshed via scripts/refresh_market_signals.py.
  • Collection statistics: the statistical scope is synchronized with tutorials/tutorial-manifest.json and the baseline check in scripts/docs_health.py.

How the Tutorials Are Built

Each tutorial is generated through AI-driven codebase analysis, then reviewed and enhanced to ensure accuracy. Its process is inspired by Tutorial-Codebase-Knowledge by The Pocket.

Use Cases

  • Developers who want to deeply understand the internal mechanisms of AI applications, agents, RAG, LLM services, and AI coding assistants.
  • Readers who need to systematically learn AI engineering topics by goal or learning path, and can use the query hub and intent map to locate content.
  • Learners who want to build AI tool servers around MCP and extend Claude's custom capabilities.
  • Technical personnel who follow open-source coding agents, Vibe-Coding ecosystem developments, and market signals.

Limitations and Not Yet Provided

  • The materials do not provide information on fees, pricing, payment models, or platform support, and it is not yet provided.
  • The materials do not provide the specific languages, license details, installation methods, or usage commands of the tutorials, and it is not yet provided.
  • The materials do not provide access channels other than the GitHub repository or specific community entry points, and it is not yet provided.

How to Contribute

The repository welcomes contributions and provides the standards for high-quality tutorials: explain the "how" and "why" in depth rather than merely stating the "what"; use examples from real codebases rather than toy demos; include visualizations (architecture diagrams, flowcharts, sequence diagrams); build complexity progressively; be production-oriented (covering deployment, monitoring, and scaling). You can suggest new tutorials or report issues by opening an Issue.

Related tools

Browse category