经 AI Skill Hub 精选评估，awesome-LangGraph — AI Agent 工作流中文教程 获评「强烈推荐」。已获得 1.8k 颗 GitHub Star，这款Agent工作流在功能完整性、社区活跃度和易用性方面表现出色，AI 评分 8.6 分，适合有一定技术背景的用户使用。
awesome-LangGraph — AI Agent 工作流中文教程 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排，将复杂的多步骤任务拆解为清晰的自动化流程，实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成，适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
awesome-LangGraph — AI Agent 工作流中文教程 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排，将复杂的多步骤任务拆解为清晰的自动化流程，实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成，适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 方式一：npm 全局安装 npm install -g awesome-langgraph # 方式二：npx 直接运行（无需安装） npx awesome-langgraph --help # 方式三：项目依赖安装 npm install awesome-langgraph # 方式四：从源码运行 git clone https://github.com/vonzosten/awesome-LangGraph cd awesome-LangGraph npm install npm start
# 命令行使用
awesome-langgraph --help
# 基本用法
awesome-langgraph [options] <input>
# Node.js 代码中使用
const awesome_langgraph = require('awesome-langgraph');
const result = await awesome_langgraph.run(options);
console.log(result);
# awesome-langgraph 配置说明 # 查看配置选项 awesome-langgraph --config-example > config.yml # 常见配置项 # output_dir: ./output # log_level: info # workers: 4 # 环境变量（覆盖配置文件） export AWESOME_LANGGRAPH_CONFIG="/path/to/config.yml"
The definitive index of frameworks, templates, and real-world projects for teams that want to build, observe, evaluate, and deploy stateful, tool-using AI agents with the LangChain + LangGraph stack.
Whether you’re prototyping your first workflow or operating production systems, this list maps the full lifecycle of agent development, from building with core libraries and integrations, to observing runs with platform tooling, evaluating quality and behavior, and deploying reliable agent applications.
What you’ll find - Core frameworks: LangChain, LangGraph, Deep Agents, and LangSmith - Resources for building, observing, evaluating, and deploying agent systems - Integrations & MCP tooling across models, vector stores, loaders, and tools - Official LangChain/LangGraph projects and prebuilt agent libraries - Community projects grouped by use case (RAG, web automation, research, finance, etc.) - Starter templates and learning resources to get productive fast
Contributions welcome—see the Contributing Guide.
---
Advanced capabilities and techniques for sophisticated AI applications
| Feature | Description |
|---|---|
| **🧠 [Long-term Memory](https://docs.langchain.com/oss/python/langchain/long-term-memory)** | Persistent memory that survives across sessions |
| **🛡️ [Guardrails](https://docs.langchain.com/oss/python/langchain/guardrails)** | Safety checks and policy enforcement for agent inputs, outputs, and tool usage |
| **🎯 [Context Engineering](https://docs.langchain.com/oss/python/langchain/context-engineering)** | Techniques for optimizing prompts and context management |
| **📋 [Structured Output](https://docs.langchain.com/oss/python/langchain/structured-output)** | Generate responses in specific formats and schemas |
| **🔗 [Model Context Protocol](https://docs.langchain.com/oss/python/langchain/mcp)** | Standardized tool integration and context sharing |
| **👥 [Human-in-the-Loop](https://docs.langchain.com/oss/python/langchain/human-in-the-loop)** | Approval workflows and interrupt-based human oversight for sensitive agent actions |
| **🤝 [Multi-agent](https://docs.langchain.com/oss/python/langchain/multi-agent)** | Coordinated systems with multiple AI agents |
| **🔍 [Retrieval](https://docs.langchain.com/oss/python/langchain/retrieval)** | Advanced document retrieval and RAG patterns |
| **⚙️ [Runtime](https://docs.langchain.com/oss/python/langchain/runtime)** | Production deployment and runtime management |
| **🔧 [Middleware](https://docs.langchain.com/oss/python/langchain/middleware)** | Custom processing layers and request/response modification |
</details>
<details> <summary><strong>▫️ LangChain Libraries ▫️</strong></summary>
| Package | Python | TypeScript | Description |
|---|---|---|---|
| **LangChain** | [langchain](https://github.com/langchain-ai/langchain/tree/master/libs/langchain) | [langchain](https://github.com/langchain-ai/langchainjs/tree/main/langchain) | Main framework with chains, agents, retrieval methods, and cognitive architecture |
| **LangChain Core** | [langchain-core](https://github.com/langchain-ai/langchain/tree/master/libs/core) | [@langchain/core](https://github.com/langchain-ai/langchainjs/tree/main/libs/langchain-core) | Base abstractions and runtime for the entire ecosystem |
| **Community** | [langchain-community](https://github.com/langchain-ai/langchain/tree/master/libs/community) | [@langchain/community](https://github.com/langchain-ai/langchainjs/tree/main/libs/langchain-community) | Third-party integrations and community contributions |
| **MCP Adapters** | [langchain-mcp-adapters](https://github.com/langchain-ai/langchain-mcp-adapters) | - | Make Anthropic MCP tools compatible with agents |
| **Text Splitters** | [langchain-text-splitters](https://github.com/langchain-ai/langchain/tree/master/libs/text-splitters) | [@langchain/textsplitters](https://github.com/langchain-ai/langchainjs/tree/main/libs/langchain-textsplitters) | Document processing and text splitting utilities |
| **Experimental** | [langchain-experimental](https://github.com/langchain-ai/langchain/tree/master/libs/experimental) | [@langchain/experimental](https://github.com/langchain-ai/langchainjs/tree/main/libs/langchain-experimental) | Beta features and experimental components |
| **CLI Tools** | [langchain-cli](https://github.com/langchain-ai/langchain/tree/master/libs/cli) | - | Command line interface for project management |
| **Legacy** | [langchain-legacy](https://github.com/langchain-ai/langchain/tree/master/libs/legacy) | - | Legacy components from pre-v1.0 (Python only) |
<details> <summary><strong>▫️ LangChain Documentation ▫️</strong></summary>
Access the official LangChain documentation across the current unified docs experience and legacy redirect URLs:
| Docs | Python | JavaScript | Notes |
|---|---|---|---|
| **Current Open Source Docs** | [Overview](https://docs.langchain.com/oss/python/langchain/overview) | [Overview](https://docs.langchain.com/oss/javascript/langchain/overview) | Current unified LangChain OSS docs on docs.langchain.com |
| **Legacy Redirects** | [Legacy Entry](https://python.langchain.com/docs/introduction/) | [Legacy Entry](https://js.langchain.com/docs/introduction/) | Older URLs that now redirect to the current overview docs |
</div>
AI-accessible documentation format for LLMs and IDEs - LangChain now exposes a unified llms.txt entrypoint on docs.langchain.com for programmatic access to the latest documentation across LangChain, LangGraph, LangSmith, and API references.
</div>
Browser control, web task automation, and data extraction
| Project | Description | GitHub Stars |
|---|---|---|
| [esinecan/agentic-ai-browser](https://github.com/esinecan/agentic-ai-browser) | Web automation agent with behavioral caching, DOM fidelity, and success pattern recording | <br> |
| [browser-use/browser-use](https://github.com/browser-use/browser-use) | Library for AI agents to control websites and automate tasks | <br> |
| [stanford-mast/blast](https://github.com/stanford-mast/blast) | High-performance serving engine for browser-augmented LLM applications with auto-scaling and OpenAI-compatible API | <br> |
| [ScrapeGraphAI/scrapecraft](https://github.com/ScrapeGraphAI/scrapecraft) | Visual editor for building scraping workflows with LangGraph, bulk scraping, and live streaming | <br> |
| [nickhawn/news-agent](https://github.com/nickhawn/news-agent) | News crawler that personalizes daily summaries with Tavily and memory | <br> |
| [hermesagent/langchain-hermes](https://github.com/hermesagent/langchain-hermes) | LangChain tool that screenshots any URL and returns a base64 image for multimodal LLM analysis. No signup required for basic use. | <br> |
</div>
Frontend applications, chat interfaces, and graphical user interfaces for AI agents
| Project | Description | GitHub Stars |
|---|---|---|
| [GaiZhenbiao/ChuanhuChatGPT](https://github.com/GaiZhenbiao/ChuanhuChatGPT) | GUI for ChatGPT/LLMs with agent support, web search, and knowledge base features | <br> |
| [CopilotKit/open-multi-agent-canvas](https://github.com/CopilotKit/open-multi-agent-canvas) | Multi-agent chat interface with travel/research examples and MCP servers | <br> |
| [teddynote-lab/LangConnect-Client](https://github.com/teddynote-lab/LangConnect-Client) | Streamlit RAG client with document management, semantic/hybrid search, and MCP integration | <br> |
🔗 LangChain - Provides integrations and composable components to streamline LLM application development. Contains agent abstractions built on top of LangGraph.
🕸️ LangGraph - The core framework for building stateful, multi-agent systems with complex workflows, collaboration, and memory management.
🧠 Deep Agents - An agent harness for building agents that can plan, decompose complex tasks, use subagents, manage large context with filesystem tools, and persist long-term memory.
🛠️ LangSmith - The platform layer for observing, evaluating, and deploying AI agents and LLM applications with tracing, prompt engineering, Agent Server deployment, sandboxes, and operational tooling.
🧩 LangSmith Fleet - A no-code platform for creating and managing AI agents from templates, connecting apps and accounts, automating routine work, and keeping humans in control with approvals and oversight.
🤝 LangChain Integrations & Partners - Third-party integrations and provider packages that extend LangChain's capabilities across the AI ecosystem. These integration packages provide standardized interfaces to work with popular AI services, databases, and tools.
---
Essential building blocks for LangChain applications
| Component | Description |
|---|---|
| **🤖 [Agents](https://docs.langchain.com/oss/python/langchain/agents)** | Decision-making systems that use LLMs to determine which actions to take |
| **🧠 [Models](https://docs.langchain.com/oss/python/langchain/models)** | Unified interfaces for LLMs and embedding models across providers |
| **💬 [Messages](https://docs.langchain.com/oss/python/langchain/messages)** | Structured communication format between components |
| **🛠️ [Tools](https://docs.langchain.com/oss/python/langchain/tools)** | External function calls and integrations for agents |
| **🧭 [Short-term Memory](https://docs.langchain.com/oss/python/langchain/short-term-memory)** | Working memory for maintaining conversation context |
| **⚡ [Streaming](https://docs.langchain.com/oss/python/langchain/streaming)** | Real-time response processing for partial results |
</div>
Third-party integrations and provider packages that extend LangChain's capabilities across the AI ecosystem. These integration packages provide standardized interfaces to work with popular AI services, databases, and tools.
<details> <summary><strong>🔸 Chat Models 🔸</strong></summary>
Language models that use message sequences as input/output for conversational AI. Support tool calling, structured output, streaming, and multimodal inputs for building sophisticated chat applications.
</details>
<details> <summary><strong>🔸 Tools & Toolkits 🔸</strong></summary>
Enable agents to interact with external systems. Define input schemas for tool calling and executing actions. Support web search, database queries, file operations, browser control, and API integrations.
</details>
<details> <summary><strong>🔸 Middleware 🔸</strong></summary>
Control and customize agent execution at every step. Add logging, retries, guardrails, human approval, rate limiting, prompt transforms, and other execution-time behavior.
</details>
<details> <summary><strong>🔸 Sandboxes 🔸</strong></summary>
Run agent-generated code in isolated execution environments. Sandboxes provide safer boundaries for shell access, filesystem operations, and code execution.
</details>
<details> <summary><strong>🔸 Checkpointers 🔸</strong></summary>
Persistence backends for LangGraph state. Save and resume agent state across interactions using in-memory, SQL, Redis, MongoDB, cloud, and other checkpoint stores.
</details>
<details> <summary><strong>🔸 Retrievers 🔸</strong></summary>
Advanced retrieval strategies that combine dense and sparse search methods. Enable sophisticated document retrieval patterns including hybrid search, reranking, and context-aware retrieval for RAG applications.
</details>
<details> <summary><strong>🔸 Text Splitters 🔸</strong></summary>
Break large documents into smaller, manageable chunks. Maintain semantic coherence while fitting within model context windows. Essential for RAG pipelines and document processing workflows.
</details>
<details> <summary><strong>🔸 Embedding Models 🔸</strong></summary>
Transform raw text into fixed-length vectors that capture semantic meaning. Enable machines to compare and search text based on meaning rather than exact words. Essential for retrieval, semantic search, and ranking workflows.
</details>
<details> <summary><strong>🔸 Vector Stores 🔸</strong></summary>
Databases optimized for storing and querying high-dimensional vectors using similarity metrics. Used with embedding models to power semantic search, retrieval, and knowledge base applications.
</details>
<details> <summary><strong>🔸 Document Loaders 🔸</strong></summary>
Integrations for ingesting data from files, websites, databases, APIs, and cloud systems. Transform diverse sources into LangChain Document objects for downstream processing.
</details>
<n></n>
---
AI Skill Hub 为第三方内容聚合平台，本页面信息基于公开数据整理，不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后，再部署至生产环境，并做好必要的安全评估。
✅ CC0 1.0 — 公共领域贡献，完全放弃版权，无任何使用限制。
AI Skill Hub 点评：awesome-LangGraph — AI Agent 工作流中文教程 的核心功能完整，质量优秀。对于自动化工程师和运维人员来说，这是一个值得纳入个人工具库的选择。建议先在非生产环境试用，再逐步推广。
| 原始名称 | awesome-LangGraph |
| 原始描述 | An index of the LangChain + LangGraph ecosystem: concepts, projects, tools, templates, and guides for LLM & multi-agent apps. |
| Topics | aiawesomeawesome-listlangchainlanggraphllmmust-have-ai |
| GitHub | https://github.com/vonzosten/awesome-LangGraph |
| License | CC0-1.0 |
| 语言 | JavaScript |
收录时间：2026-05-22 · 更新时间：2026-05-22 · License：CC0-1.0 · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型，复制安装指令后粘贴到对应客户端