agent-craft and AgentGuide
These are **complements** — Annyfee/agent-craft provides systematized runnable code examples for building agents with LangChain/RAG/LangGraph/MCP, while adongwanai/AgentGuide offers comprehensive development guidance, advanced RAG patterns, and interview preparation to understand the conceptual and practical foundations needed to effectively use those tools.
About agent-craft
Annyfee/agent-craft
AI Agent 教学仓库 | 系统化 LangChain、RAG、LangGraph、MCP 全栈实战代码 | 万字博客详解 | 开源可运行示例 | 从零构建智能体
Implements modular agent capabilities through progressive scaffolding: LangChain fundamentals, RAG with vector stores (FAISS/Chroma) and reranking, LangGraph for stateful agentic workflows with LangSmith debugging, MCP server/client patterns using Stdio and HTTP transports, and multi-agent orchestration via Swarm protocol with Claude API. Each of 13+ modules features standalone runnable code with companion blogs explaining design patterns, plus Streamlit UI bindings and deployment examples (Ollama, LM Studio, LangServe).
About AgentGuide
adongwanai/AgentGuide
https://adongwanai.github.io/AgentGuide | AI Agent开发指南 | LangGraph实战 | 高级RAG | 转行大模型 | 大模型面试 | 算法工程师 | 面试题库 | 强化学习｜数据合成
# Technical Summary Provides a systematic, job-market-oriented learning pathway covering the full AI Agent tech stack—from LangGraph/LangChain frameworks and advanced RAG techniques (GraphRAG, reranking) to multi-agent orchestration and context engineering strategies. Distinguishes between algorithm (research-focused, model optimization via SFT/LoRA/RLHF) and engineering (production systems, retrieval pipelines, deployment) career tracks, with curated open-source projects and 1000+ interview problems tied to job preparation. The guide integrates vector databases, embedding models, and workflow orchestration tools (Dify, n8n) while emphasizing portfolio-building through complete end-to-end systems—RAG agents, multi-agent workflows, web automation—rather than isolated demos, designed for both researchers publishing papers and engineers shipping production services.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work