[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-langchain-ai--langgraphjs":3,"tool-langchain-ai--langgraphjs":64},[4,17,27,35,43,56],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":16},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,3,"2026-04-05T11:01:52",[13,14,15],"开发框架","图像","Agent","ready",{"id":18,"name":19,"github_repo":20,"description_zh":21,"stars":22,"difficulty_score":23,"last_commit_at":24,"category_tags":25,"status":16},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",140436,2,"2026-04-05T23:32:43",[13,15,26],"语言模型",{"id":28,"name":29,"github_repo":30,"description_zh":31,"stars":32,"difficulty_score":23,"last_commit_at":33,"category_tags":34,"status":16},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",107662,"2026-04-03T11:11:01",[13,14,15],{"id":36,"name":37,"github_repo":38,"description_zh":39,"stars":40,"difficulty_score":23,"last_commit_at":41,"category_tags":42,"status":16},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[13,26],{"id":44,"name":45,"github_repo":46,"description_zh":47,"stars":48,"difficulty_score":23,"last_commit_at":49,"category_tags":50,"status":16},2268,"ML-For-Beginners","microsoft\u002FML-For-Beginners","ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程，旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周，包含 26 节精炼课程和 52 道配套测验，内容涵盖从基础概念到实际应用的完整流程，有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。\n\n无论是希望转型的开发者、需要补充算法背景的研究人员，还是对人工智能充满好奇的普通爱好者，都能从中受益。课程不仅提供了清晰的理论讲解，还强调动手实践，让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持，通过自动化机制提供了包括简体中文在内的 50 多种语言版本，极大地降低了全球不同背景用户的学习门槛。此外，项目采用开源协作模式，社区活跃且内容持续更新，确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路，ML-For-Beginners 将是理想的起点。",84991,"2026-04-05T10:45:23",[14,51,52,53,15,54,26,13,55],"数据工具","视频","插件","其他","音频",{"id":57,"name":58,"github_repo":59,"description_zh":60,"stars":61,"difficulty_score":10,"last_commit_at":62,"category_tags":63,"status":16},3128,"ragflow","infiniflow\u002Fragflow","RAGFlow 是一款领先的开源检索增强生成（RAG）引擎，旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体（Agent）能力相结合，不仅支持从各类文档中高效提取知识，还能让模型基于这些知识进行逻辑推理和任务执行。\n\n在大模型应用中，幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构（如表格、图表及混合排版），显著提升了信息检索的准确度，从而有效减少模型“胡编乱造”的现象，确保回答既有据可依又具备时效性。其内置的智能体机制更进一步，使系统不仅能回答问题，还能自主规划步骤解决复杂问题。\n\n这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统，还是致力于探索大模型在垂直领域落地的创新者，都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口，既降低了非算法背景用户的上手门槛，也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目，它正成为连接通用大模型与行业专有知识之间的重要桥梁。",77062,"2026-04-04T04:44:48",[15,14,13,26,54],{"id":65,"github_repo":66,"name":67,"description_en":68,"description_zh":69,"ai_summary_zh":69,"readme_en":70,"readme_zh":71,"quickstart_zh":72,"use_case_zh":73,"hero_image_url":74,"owner_login":75,"owner_name":76,"owner_avatar_url":77,"owner_bio":78,"owner_company":79,"owner_location":79,"owner_email":80,"owner_twitter":76,"owner_website":81,"owner_url":82,"languages":83,"stars":111,"forks":112,"last_commit_at":113,"license":114,"difficulty_score":23,"env_os":115,"env_gpu":116,"env_ram":116,"env_deps":117,"category_tags":125,"github_topics":126,"view_count":23,"oss_zip_url":79,"oss_zip_packed_at":79,"status":16,"created_at":134,"updated_at":135,"faqs":136,"releases":165},2384,"langchain-ai\u002Flanggraphjs","langgraphjs","Framework to build resilient language agents as graphs.","langgraphjs 是一个专为 JavaScript 和 TypeScript 开发者设计的开源框架，旨在通过“图”的结构构建高可靠性、可控的智能体（AI Agents）。它解决了传统大模型应用在处理复杂任务时难以维持长期记忆、缺乏流程控制以及无法灵活介入人工审核等痛点，让开发者能够轻松编排多步骤工作流。\n\n该工具特别适合需要深度定制 AI 行为的后端工程师、全栈开发者及架构师。不同于仅提供简单组件库的方案，langgraphjs 提供了底层的原语支持，允许用户自由定义智能体的状态流转与交互逻辑，从而构建出适应特定业务场景的多智能体系统。其核心技术亮点包括原生支持长短期记忆持久化、内置“人在回路”（Human-in-the-loop）机制以确保关键决策的安全性，以及提供细粒度的实时流式输出，让用户能清晰观察智能体的推理过程。目前，该框架已被 Uber、Replit 等企业用于生产环境，是打造健壮且可扩展 AI 应用的理想选择。","# 🦜🕸️LangGraph.js\n\n[![Docs](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fdocs-latest-blue)](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002F)\n![Version](https:\u002F\u002Fimg.shields.io\u002Fnpm\u002Fv\u002F@langchain\u002Flanggraph?logo=npm)  \n[![Downloads](https:\u002F\u002Fimg.shields.io\u002Fnpm\u002Fdm\u002F@langchain\u002Flanggraph)](https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002F@langchain\u002Flanggraph)\n[![Open Issues](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues-raw\u002Flangchain-ai\u002Flanggraphjs)](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fissues)\n\n> [!NOTE]\n> Looking for the Python version? See the [Python repo](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraph) and the [Python docs](https:\u002F\u002Fdocs.langchain.com\u002Foss\u002Fpython\u002Flanggraph\u002Foverview).\n\nLangGraph — used by Replit, Uber, LinkedIn, GitLab and more — is a low-level orchestration framework for building controllable agents. While langchain provides integrations and composable components to streamline LLM application development, the LangGraph library enables agent orchestration — offering customizable architectures, long-term memory, and human-in-the-loop to reliably handle complex tasks.\n\n```bash\nnpm install @langchain\u002Flanggraph @langchain\u002Fcore\n```\n\nTo learn more about how to use LangGraph, check out [the docs](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002F). We show a simple example below of how to create a ReAct agent.\n\n```ts\n\u002F\u002F npm install @langchain-anthropic\nimport { createReactAgent, tool } from \"langchain\";\nimport { ChatAnthropic } from \"@langchain\u002Fanthropic\";\n\nimport { z } from \"zod\";\n\nconst search = tool(\n  async ({ query }) => {\n    if (\n      query.toLowerCase().includes(\"sf\") ||\n      query.toLowerCase().includes(\"san francisco\")\n    ) {\n      return \"It's 60 degrees and foggy.\";\n    }\n    return \"It's 90 degrees and sunny.\";\n  },\n  {\n    name: \"search\",\n    description: \"Call to surf the web.\",\n    schema: z.object({\n      query: z.string().describe(\"The query to use in your search.\"),\n    }),\n  }\n);\n\nconst model = new ChatAnthropic({\n  model: \"claude-3-7-sonnet-latest\",\n});\n\nconst agent = createReactAgent({\n  llm: model,\n  tools: [search],\n});\n\nconst result = await agent.invoke({\n  messages: [\n    {\n      role: \"user\",\n      content: \"what is the weather in sf\",\n    },\n  ],\n});\n```\n\n## Full-stack Quickstart\n\nGet started quickly by building a full-stack LangGraph application using the [`create-agent-chat-app`](https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002Fcreate-agent-chat-app) CLI:\n\n```bash\nnpx create-agent-chat-app@latest\n```\n\nThe CLI sets up a chat interface and helps you configure your application, including:\n\n- 🧠 Choice of 4 prebuilt agents (ReAct, Memory, Research, Retrieval)\n- 🌐 Frontend framework (Next.js or Vite)\n- 📦 Package manager (`npm`, `yarn`, or `pnpm`)\n\n## Why use LangGraph?\n\nLangGraph is built for developers who want to build powerful, adaptable AI agents. Developers choose LangGraph for:\n\n- **Reliability and controllability.** Steer agent actions with moderation checks and human-in-the-loop approvals. LangGraph persists context for long-running workflows, keeping your agents on course.\n- **Low-level and extensible.** Build custom agents with fully descriptive, low-level primitives – free from rigid abstractions that limit customization. Design scalable multi-agent systems, with each agent serving a specific role tailored to your use case.\n- **First-class streaming support.** With token-by-token streaming and streaming of intermediate steps, LangGraph gives users clear visibility into agent reasoning and actions as they unfold in real time.\n\nLangGraph is trusted in production and powering agents for companies like:\n\n- [Klarna](https:\u002F\u002Fblog.langchain.dev\u002Fcustomers-klarna\u002F): Customer support bot for 85 million active users\n- [Elastic](https:\u002F\u002Fwww.elastic.co\u002Fblog\u002Felastic-security-generative-ai-features): Security AI assistant for threat detection\n- [Uber](https:\u002F\u002Fdpe.org\u002Fsessions\u002Fty-smith-adam-huda\u002Fthis-year-in-ubers-ai-driven-developer-productivity-revolution\u002F): Automated unit test generation\n- [Replit](https:\u002F\u002Fwww.langchain.com\u002Fbreakoutagents\u002Freplit): Code generation\n- And many more ([see list here](https:\u002F\u002Fwww.langchain.com\u002Fbuilt-with-langgraph))\n\n## LangGraph’s ecosystem\n\nWhile LangGraph can be used standalone, it also integrates seamlessly with any LangChain product, giving developers a full suite of tools for building agents. To improve your LLM application development, pair LangGraph with:\n\n- [LangSmith](http:\u002F\u002Fwww.langchain.com\u002Flangsmith) — Helpful for agent evals and observability. Debug poor-performing LLM app runs, evaluate agent trajectories, gain visibility in production, and improve performance over time.\n- [LangGraph Platform](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Fconcepts\u002F#langgraph-platform) — Deploy and scale agents effortlessly with a purpose-built deployment platform for long running, stateful workflows. Discover, reuse, configure, and share agents across teams — and iterate quickly with visual prototyping in [LangGraph Studio](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Fconcepts\u002Flanggraph_studio\u002F).\n\n## Pairing with LangGraph Platform\n\nWhile LangGraph is our open-source agent orchestration framework, enterprises that need scalable agent deployment can benefit from [LangGraph Platform](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Fconcepts\u002Flanggraph_platform\u002F).\n\nLangGraph Platform can help engineering teams:\n\n- **Accelerate agent development**: Quickly create agent UXs with configurable templates and [LangGraph Studio](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Fconcepts\u002Flanggraph_studio\u002F) for visualizing and debugging agent interactions.\n- **Deploy seamlessly**: We handle the complexity of deploying your agent. LangGraph Platform includes robust APIs for memory, threads, and cron jobs plus auto-scaling task queues & servers.\n- **Centralize agent management & reusability**: Discover, reuse, and manage agents across the organization. Business users can also modify agents without coding.\n\n## Additional resources\n\n- [LangChain Forum](https:\u002F\u002Fforum.langchain.com\u002F): Connect with the community and share all of your technical questions, ideas, and feedback.\n- [LangChain Academy](https:\u002F\u002Facademy.langchain.com\u002Fcourses\u002Fintro-to-langgraph): Learn the basics of LangGraph in our free, structured course.\n- [Tutorials](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Ftutorials\u002F): Simple walkthroughs with guided examples on getting started with LangGraph.\n- [Templates](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Fconcepts\u002Ftemplate_applications\u002F): Pre-built reference apps for common agentic workflows (e.g. ReAct agent, memory, retrieval etc.) that can be cloned and adapted.\n- [How-to Guides](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Fhow-tos\u002F): Quick, actionable code snippets for topics such as streaming, adding memory & persistence, and design patterns (e.g. branching, subgraphs, etc.).\n- [API Reference](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Freference\u002F): Detailed reference on core classes, methods, how to use the graph and checkpointing APIs, and higher-level prebuilt components.\n- [Built with LangGraph](https:\u002F\u002Fwww.langchain.com\u002Fbuilt-with-langgraph): Hear how industry leaders use LangGraph to ship powerful, production-ready AI applications.\n\n## Acknowledgements\n\nLangGraph is inspired by [Pregel](https:\u002F\u002Fresearch.google\u002Fpubs\u002Fpub37252\u002F) and [Apache Beam](https:\u002F\u002Fbeam.apache.org\u002F). The public interface draws inspiration from [NetworkX](https:\u002F\u002Fnetworkx.org\u002Fdocumentation\u002Flatest\u002F). LangGraph is built by LangChain Inc, the creators of LangChain, but can be used without LangChain.\n","# 🦜🕸️LangGraph.js\n\n[![文档](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fdocs-latest-blue)](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002F)\n![版本](https:\u002F\u002Fimg.shields.io\u002Fnpm\u002Fv\u002F@langchain\u002Flanggraph?logo=npm)  \n[![下载量](https:\u002F\u002Fimg.shields.io\u002Fnpm\u002Fdm\u002F@langchain\u002Flanggraph)](https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002F@langchain\u002Flanggraph)\n[![未解决问题](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues-raw\u002Flangchain-ai\u002Flanggraphjs)](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fissues)\n\n> [!注意]\n> 您在寻找 Python 版本吗？请参阅 [Python 仓库](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraph) 和 [Python 文档](https:\u002F\u002Fdocs.langchain.com\u002Foss\u002Fpython\u002Flanggraph\u002Foverview)。\n\nLangGraph — 被 Replit、Uber、LinkedIn、GitLab 等公司使用 — 是一个用于构建可控智能体的底层编排框架。虽然 LangChain 提供了集成和可组合组件来简化 LLM 应用程序开发，但 LangGraph 库则专注于智能体编排，提供可定制的架构、长期记忆以及人机协作功能，从而可靠地处理复杂任务。\n\n```bash\nnpm install @langchain\u002Flanggraph @langchain\u002Fcore\n```\n\n要了解更多关于如何使用 LangGraph 的信息，请查看 [文档](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002F)。下面是一个简单的示例，展示如何创建一个 ReAct 智能体。\n\n```ts\n\u002F\u002F npm install @langchain-anthropic\nimport { createReactAgent, tool } from \"langchain\";\nimport { ChatAnthropic } from \"@langchain\u002Fanthropic\";\n\nimport { z } from \"zod\";\n\nconst search = tool(\n  async ({ query }) => {\n    if (\n      query.toLowerCase().includes(\"sf\") ||\n      query.toLowerCase().includes(\"san francisco\")\n    ) {\n      return \"现在是60华氏度，有雾。\";\n    }\n    return \"现在是90华氏度，晴朗。\";\n  },\n  {\n    name: \"search\",\n    description: \"用于在网络上搜索的工具。\",\n    schema: z.object({\n      query: z.string().describe(\"您要在搜索中使用的查询。\"),\n    }),\n  }\n);\n\nconst model = new ChatAnthropic({\n  model: \"claude-3-7-sonnet-latest\",\n});\n\nconst agent = createReactAgent({\n  llm: model,\n  tools: [search],\n});\n\nconst result = await agent.invoke({\n  messages: [\n    {\n      role: \"user\",\n      content: \"旧金山的天气怎么样\",\n    },\n  ],\n});\n```\n\n## 全栈快速入门\n\n通过使用 [`create-agent-chat-app`](https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002Fcreate-agent-chat-app) CLI 快速构建一个全栈 LangGraph 应用程序：\n\n```bash\nnpx create-agent-chat-app@latest\n```\n\n该 CLI 会设置聊天界面，并帮助您配置应用程序，包括：\n\n- 🧠 四种预建智能体可供选择（ReAct、Memory、Research、Retrieval）\n- 🌐 前端框架（Next.js 或 Vite）\n- 📦 包管理器（`npm`、`yarn` 或 `pnpm`）\n\n## 为什么使用 LangGraph？\n\nLangGraph 专为希望构建强大且适应性强的 AI 智能体的开发者而设计。开发者选择 LangGraph 的原因在于：\n\n- **可靠性和可控性。** 通过审核机制和人工审批来引导智能体的行为。LangGraph 可以持久化上下文信息，支持长时间运行的工作流，确保您的智能体始终按计划执行。\n- **低层级且可扩展。** 使用完全描述性的低层级原语构建自定义智能体，摆脱限制自定义的僵化抽象。您可以设计可扩展的多智能体系统，每个智能体根据您的应用场景承担特定角色。\n- **一流的流式传输支持。** LangGraph 支持逐个 token 的流式传输以及中间步骤的流式输出，让用户能够实时清晰地了解智能体的推理过程和行动。\n\nLangGraph 已被广泛应用于生产环境，并为以下公司提供了智能体支持：\n\n- [Klarna](https:\u002F\u002Fblog.langchain.dev\u002Fcustomers-klarna\u002F)：为8500万活跃用户提供客服机器人\n- [Elastic](https:\u002F\u002Fwww.elastic.co\u002Fblog\u002Felastic-security-generative-ai-features)：用于威胁检测的安全 AI 助手\n- [Uber](https:\u002F\u002Fdpe.org\u002Fsessions\u002Fty-smith-adam-huda\u002Fthis-year-in-ubers-ai-driven-developer-productivity-revolution\u002F)：自动生成单元测试\n- [Replit](https:\u002F\u002Fwww.langchain.com\u002Fbreakoutagents\u002Freplit)：代码生成\n- 以及更多公司（[查看完整列表](https:\u002F\u002Fwww.langchain.com\u002Fbuilt-with-langgraph)）\n\n## LangGraph 的生态系统\n\n尽管 LangGraph 可以独立使用，但它也能与任何 LangChain 产品无缝集成，为开发者提供一套完整的智能体构建工具。为了提升您的 LLM 应用程序开发效率，可以将 LangGraph 与以下产品结合使用：\n\n- [LangSmith](http:\u002F\u002Fwww.langchain.com\u002Flangsmith) — 非常适合智能体评估和可观ability。您可以调试表现不佳的 LLM 应用程序运行情况，评估智能体的决策路径，在生产环境中获得透明度，并持续改进性能。\n- [LangGraph 平台](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Fconcepts\u002F#langgraph-platform) — 通过专门构建的部署平台，轻松部署和扩展长期运行的状态化工作流。您可以在团队之间发现、重用、配置和共享智能体，并借助 [LangGraph Studio](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Fconcepts\u002Flanggraph_studio\u002F) 进行可视化原型设计，快速迭代。\n\n## 与 LangGraph 平台的结合使用\n\n虽然 LangGraph 是我们的开源智能体编排框架，但对于需要规模化部署智能体的企业而言，[LangGraph 平台](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Fconcepts\u002Flanggraph_platform\u002F) 将带来显著优势。\n\nLangGraph 平台可以帮助工程团队：\n\n- **加速智能体开发**：利用可配置模板和 [LangGraph Studio](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Fconcepts\u002Flanggraph_studio\u002F) 快速创建智能体用户界面，实现智能体交互的可视化和调试。\n- **无缝部署**：我们负责处理智能体部署的复杂性。LangGraph 平台包含强大的内存、线程和定时任务 API，以及自动扩展的任务队列和服务器。\n- **集中管理和智能体复用**：在整个组织内发现、重用并管理智能体。业务人员无需编写代码即可修改智能体。\n\n## 其他资源\n\n- [LangChain 论坛](https:\u002F\u002Fforum.langchain.com\u002F)：与社区交流，分享您的技术问题、想法和反馈。\n- [LangChain 学院](https:\u002F\u002Facademy.langchain.com\u002Fcourses\u002Fintro-to-langgraph)：通过我们免费的结构化课程学习 LangGraph 的基础知识。\n- [教程](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Ftutorials\u002F)：包含引导式示例的简单入门指南，帮助您快速上手 LangGraph。\n- [模板](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Fconcepts\u002Ftemplate_applications\u002F)：针对常见代理工作流（如 ReAct 代理、记忆、检索等）的预构建参考应用，可直接克隆并进行定制。\n- [操作指南](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Fhow-tos\u002F)：涵盖流式处理、添加记忆与持久化、设计模式（如分支、子图等）等主题的快速、可执行代码片段。\n- [API 参考](https:\u002F\u002Flangchain-ai.github.io\u002Flanggraphjs\u002Freference\u002F)：详细介绍了核心类、方法，以及如何使用图和检查点 API，还有更高层次的预构建组件。\n- [由 LangGraph 构建](https:\u002F\u002Fwww.langchain.com\u002Fbuilt-with-langgraph)：了解行业领军企业如何利用 LangGraph 打造功能强大、可投入生产的 AI 应用。\n\n## 致谢\n\nLangGraph 的灵感来源于 [Pregel](https:\u002F\u002Fresearch.google\u002Fpubs\u002Fpub37252\u002F) 和 [Apache Beam](https:\u002F\u002Fbeam.apache.org\u002F)。其公共接口的设计则受到 [NetworkX](https:\u002F\u002Fnetworkx.org\u002Fdocumentation\u002Flatest\u002F) 的启发。LangGraph 由 LangChain Inc. 开发，该公司也是 LangChain 的创建者，但 LangGraph 无需依赖 LangChain 即可使用。","# LangGraph.js 快速上手指南\n\nLangGraph.js 是一个用于构建可控、有状态 AI 智能体（Agent）的低级编排框架。它支持长期记忆、人机协作（Human-in-the-loop）以及复杂的任务流处理，被 Replit、Uber、LinkedIn 等公司广泛应用于生产环境。\n\n## 环境准备\n\n在开始之前，请确保您的开发环境满足以下要求：\n\n*   **Node.js**: 建议安装 LTS 版本（v18 或更高）。\n*   **包管理器**: `npm`、`yarn` 或 `pnpm`。\n*   **前置知识**: 熟悉 TypeScript\u002FJavaScript 基础及 LangChain 核心概念（可选，但推荐）。\n\n> **提示**：国内开发者若遇到网络延迟，可配置国内镜像源加速依赖下载：\n> ```bash\n> npm config set registry https:\u002F\u002Fregistry.npmmirror.com\n> ```\n\n## 安装步骤\n\n通过 npm 安装核心库及 LangChain 核心依赖：\n\n```bash\nnpm install @langchain\u002Flanggraph @langchain\u002Fcore\n```\n\n如果您计划使用特定的模型提供商（如示例中的 Anthropic），还需安装对应的集成包：\n\n```bash\nnpm install @langchain\u002Fanthropic\n```\n\n此外，您还需要安装用于定义工具 Schema 的验证库：\n\n```bash\nnpm install zod\n```\n\n## 基本使用\n\n以下是一个最简示例，展示如何创建一个具备搜索能力的 **ReAct 智能体**。该智能体能根据用户输入调用自定义工具，并返回结果。\n\n### 代码示例\n\n创建文件 `agent.ts` 并填入以下代码：\n\n```ts\n\u002F\u002F npm install @langchain-anthropic\nimport { createReactAgent, tool } from \"langchain\";\nimport { ChatAnthropic } from \"@langchain\u002Fanthropic\";\n\nimport { z } from \"zod\";\n\n\u002F\u002F 1. 定义一个自定义工具 (search)\nconst search = tool(\n  async ({ query }) => {\n    if (\n      query.toLowerCase().includes(\"sf\") ||\n      query.toLowerCase().includes(\"san francisco\")\n    ) {\n      return \"It's 60 degrees and foggy.\";\n    }\n    return \"It's 90 degrees and sunny.\";\n  },\n  {\n    name: \"search\",\n    description: \"Call to surf the web.\",\n    schema: z.object({\n      query: z.string().describe(\"The query to use in your search.\"),\n    }),\n  }\n);\n\n\u002F\u002F 2. 初始化大语言模型\nconst model = new ChatAnthropic({\n  model: \"claude-3-7-sonnet-latest\",\n  \u002F\u002F 请确保在环境变量中设置了 ANTHROPIC_API_KEY\n});\n\n\u002F\u002F 3. 创建 ReAct 智能体\nconst agent = createReactAgent({\n  llm: model,\n  tools: [search],\n});\n\n\u002F\u002F 4. 调用智能体\nconst result = await agent.invoke({\n  messages: [\n    {\n      role: \"user\",\n      content: \"what is the weather in sf\",\n    },\n  ],\n});\n\nconsole.log(result.messages[result.messages.length - 1].content);\n```\n\n### 运行说明\n\n1.  确保已设置环境变量 `ANTHROPIC_API_KEY`。\n2.  使用 TypeScript 运行环境（如 `tsx` 或编译后运行）执行上述文件：\n    ```bash\n    npx tsx agent.ts\n    ```\n3.  智能体将自动分析意图，调用 `search` 工具查询旧金山天气，并生成最终回答。\n\n---\n\n**进阶启动**：若想快速构建包含前端界面的全栈应用，可使用官方 CLI 一键生成项目：\n```bash\nnpx create-agent-chat-app@latest\n```","某电商平台的开发团队正在构建一个能处理复杂售后流程（如退货、换货、赔偿协商）的智能客服 Agent，该流程需多次调用内部订单系统并涉及人工审核环节。\n\n### 没有 langgraphjs 时\n- **流程失控**：基于简单线性链的逻辑难以处理多轮对话中的状态跳转，Agent 常在等待用户补充信息或调用工具后“迷失方向”，导致死循环或错误终止。\n- **缺乏人工干预**：遇到高金额赔偿等敏感场景时，无法优雅地暂停程序等待人工审批，要么盲目自动执行风险操作，要么直接切断服务体验。\n- **上下文丢失**：长周期的售后处理中，Agent 难以持久化存储中间状态，一旦会话超时或中断，之前的调查进度全部清零，用户需重复陈述问题。\n- **调试黑盒**：当 Agent 做出错误决策时，开发者无法清晰追踪其推理路径和工具调用的具体顺序，排查问题如同大海捞针。\n\n### 使用 langgraphjs 后\n- **图式编排可控**：利用有向图结构精确定义状态流转，无论用户如何跳跃提问，Agent 都能依据预设节点准确执行“查询 - 判断 - 行动”逻辑，杜绝流程混乱。\n- **原生支持人机协作**：通过内置的“人类介入”节点，在触发高风险操作时自动挂起任务，待客服人员确认后再继续执行，兼顾自动化效率与业务安全。\n- **长效记忆持久化**：借助检查点机制自动保存每一步的状态快照，即使跨天处理或系统重启，Agent 也能瞬间恢复上下文，无缝接续未完成的工单。\n- **全链路可观测**：提供细粒度的流式输出和中间步骤可视化，开发者能实时看到 Agent 的思考过程和工具参数，快速定位并优化异常行为。\n\nlanggraphjs 将原本脆弱脆弱的线性脚本升级为具备状态记忆、人工协同能力且逻辑严密的工业化智能体，让复杂业务自动化真正落地生产环境。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Flangchain-ai_langgraphjs_38958c14.png","langchain-ai","LangChain","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Flangchain-ai_8e6aaeef.png","",null,"support@langchain.dev","https:\u002F\u002Fwww.langchain.com","https:\u002F\u002Fgithub.com\u002Flangchain-ai",[84,88,92,96,100,103,107],{"name":85,"color":86,"percentage":87},"TypeScript","#3178c6",98,{"name":89,"color":90,"percentage":91},"JavaScript","#f1e05a",0.8,{"name":93,"color":94,"percentage":95},"Svelte","#ff3e00",0.7,{"name":97,"color":98,"percentage":99},"Python","#3572A5",0.2,{"name":101,"color":102,"percentage":99},"CSS","#663399",{"name":104,"color":105,"percentage":106},"Shell","#89e051",0.1,{"name":108,"color":109,"percentage":110},"HTML","#e34c26",0,2732,439,"2026-04-02T21:23:07","MIT","未说明 (基于 Node.js，通常支持 Linux, macOS, Windows)","未说明",{"notes":118,"python":119,"dependencies":120},"这是一个 JavaScript\u002FTypeScript 库，通过 npm 安装。不需要 Python 环境。示例代码展示了与 Anthropic 模型的集成，实际运行需配置对应的大模型 API Key。可使用 npx create-agent-chat-app@latest 快速搭建全栈应用，支持 Next.js 或 Vite 前端框架。","不需要 (这是 JavaScript\u002FTypeScript 库)",[121,122,123,124],"@langchain\u002Flanggraph","@langchain\u002Fcore","@langchain\u002Fanthropic (示例依赖)","zod",[15,26,14,13],[127,128,129,130,131,132,133],"agents","ai","typescript","artificial-intelligence","generative-ai","llm","node","2026-03-27T02:49:30.150509","2026-04-06T11:31:12.531922",[137,142,147,152,157,161],{"id":138,"question_zh":139,"answer_zh":140,"source_url":141},10967,"为什么在类中构建图时，LangGraph Studio 不显示 Chat 选项卡？","这是一个已知的限制或 Bug。当图（Graph）是在类内部构建并导出时，LangGraph Studio 可能无法正确识别消息键（messages key），导致 Chat 选项卡不显示。错误提示通常为 \"Create a graph with a messages key to chat with\"。目前社区反馈即使更新到最新版本（如 @langchain\u002Flanggraph ^1.2.2）该问题仍然存在。建议暂时避免在类封装中直接导出编译后的图，或者确保图的定义和导出方式符合 Studio 的静态分析要求（例如直接在模块顶层定义并导出 StateGraph）。","https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fissues\u002F1722",{"id":143,"question_zh":144,"answer_zh":145,"source_url":146},10968,"如何在浏览器或 React Native 环境中使用 LangGraph，避免 async_hooks 报错？","在浏览器或 React Native 等不支持 Node.js `async_hooks` 模块的环境中，直接导入主包会报错 \"AsyncLocalStorage is not a constructor\"。解决方案是改用专门针对 Web 环境的入口文件：将导入路径从 `@langchain\u002Flanggraph` 更改为 `@langchain\u002Flanggraph\u002Fweb`。此外，确保在所有配置传递过程中都兼容 Web 环境。对于预构建的代理（如 supervisor agent），也需要确认其内部是否使用了 Web 兼容的导入方式，否则可能需要等待官方发布完全支持 Web 的预构建版本或自行通过 Polyfill 处理（但 Polyfill 方案在某些复杂场景下可能仍不稳定）。","https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fissues\u002F81",{"id":148,"question_zh":149,"answer_zh":150,"source_url":151},10969,"使用 ChatAnthropic 模型配合工具（Tools）时，streamEvents 流式输出为何包含工具输入参数且格式混乱？","当使用 `ChatAnthropic` 并绑定工具时，`streamEvents` 会在 `on_chat_model_stream` 事件中同时输出文本流和工具输入参数（JSON 格式），导致最终展示给用户的文本中混入了类似 `{\"a\": 2, \"b\": 3}` 的原始数据。这是因为当前实现中，绑定工具不会自动过滤流中的工具块。目前的变通方法是需要在应用层手动解析流事件，区分文本内容和工具调用参数，并在渲染给用户前过滤掉工具输入的 JSON 部分。官方建议如果有关于如何正确流式传输最终结果的特定需求，可以在 langchainjs 仓库中提交更详细的问题以寻求针对性的代码示例。","https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fissues\u002F253",{"id":153,"question_zh":154,"answer_zh":155,"source_url":156},10970,"文档中的 StateGraph TypeScript 示例代码为何报错？","文档中某些旧示例直接使用普通对象定义 schema 并传递给 `StateGraph` 构造函数（如 `channels: schema`），这在较新版本的 TypeScript 类型定义中会报错，因为类型不匹配。正确的做法是使用 `Annotation.Root` 或 `StateSchema` 来明确定义状态结构，或者确保传入的对象符合 `StateGraphArgs` 中 `channels` 的严格类型要求。如果遇到类型错误，请检查 `@langchain\u002Flanggraph` 的版本，并参考最新的官方文档中关于定义 State Schema 的部分，通常需要使用 `Annotation` 辅助函数来声明带有 reducer 和 default 值的通道。","https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fissues\u002F168",{"id":158,"question_zh":159,"answer_zh":160,"source_url":151},10971,"如何在 LangGraph JS 中正确使用 v2 Schema 进行流式传输？","在使用 v2 Schema 进行流式传输（如 `streamEvents`）时，如果遇到问题，首先确保模型和工具的兼容性。例如，使用 `ChatOpenAI` 通常能正常工作，但切换到 `ChatAnthropic` 时可能需要额外配置。关键点包括：1. 确保状态定义使用了正确的 v2 语法（即通过 `Annotation` 定义 channels）；2. 检查模型是否支持流式工具调用；3. 如果使用工具，注意流事件中可能包含工具调用的原始数据，需要前端进行清洗。如果在迁移旧代码时遇到困难，可以参考社区中关于从 v1 迁移到 v2 的讨论，特别是关于 `StateGraphArgs` 定义的变更。",{"id":162,"question_zh":163,"answer_zh":164,"source_url":146},10972,"在 React Native 中使用 LangGraph Supervisor Agent 遇到构建失败怎么办？","在 React Native 中使用 `@langchain\u002Flanggraph-supervisor` 时，由于其内部依赖 `AsyncLocalStorage` (来自 `async_hooks`)，会导致构建失败。尝试使用通用的 async local storage polyfill 可能无法完全解决问题。目前的建议是：1. 检查是否有更新的包版本专门支持移动端\u002FWeb；2. 如果没有，可能需要暂时避免在 React Native 中直接使用预构建的 Supervisor Agent，而是手动实现类似逻辑，确保不使用 `async_hooks`；3. 关注官方是否发布 `web` 版本的 supervisor 包（类似于主包的 `\u002Fweb` 入口），这将是最彻底的解决方案。",[166,171,176,181,186,191,196,201,206,211,216,221,226,231,236,241,246,251,256,261],{"id":167,"version":168,"summary_zh":169,"released_at":170},53440,"@langchain\u002Flanggraph-sdk@1.8.4","### 补丁变更\n\n-   [#2263](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2263) [`936b48b`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F936b48b2807687d3fa5dd7aa480ebcc2ad3ffccf) 感谢 [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - 恢复人在回路中断负载中的已弃用 snake_case 别名，同时保留较新的 camelCase 字段，以便旧版应用可以在不破坏中断处理的情况下迁移到 `@langchain\u002Freact`。","2026-04-02T04:34:44",{"id":172,"version":173,"summary_zh":174,"released_at":175},53441,"@langchain\u002Fangular@0.3.2","### 补丁变更\n\n-   [#2253](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2253) [`76a0a94`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F76a0a94ff5eb87419e33dc3f9911a1b12ef1d6d7) 感谢 [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - Angular 事件回调测试\n\n-   更新了依赖项 \\[[`d9d807e`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002Fd9d807ebb0398a43a07412fb034a65fc598c0731)]:\n    -   @langchain\u002Flanggraph-sdk@1.8.3\n","2026-04-01T01:16:31",{"id":177,"version":178,"summary_zh":179,"released_at":180},53442,"@langchain\u002Fvue@0.3.2","### 补丁变更\n\n-   [#2252](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2252) [`6a47a82`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F6a47a826ee4c15cccedd289c267ef30eb2c64db5) 感谢 [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - 文档(vue): 修复示例中的模板引用用法\n\n-   更新了依赖项 \\[[`d9d807e`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002Fd9d807ebb0398a43a07412fb034a65fc598c0731)]:\n    -   @langchain\u002Flanggraph-sdk@1.8.3\n","2026-04-01T01:16:28",{"id":182,"version":183,"summary_zh":184,"released_at":185},53443,"@langchain\u002Flanggraph-sdk@1.8.3","### 补丁变更\n\n-   [#2204](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2204) [`d9d807e`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002Fd9d807ebb0398a43a07412fb034a65fc598c0731) 感谢 [@brydar](https:\u002F\u002Fgithub.com\u002Fbrydar)! - 修复（SDK）：在 StreamManager 中累积并行中断\n","2026-04-01T01:16:25",{"id":187,"version":188,"summary_zh":189,"released_at":190},53444,"@langchain\u002Freact@0.2.3","### 补丁变更\n\n-   [#2250](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2250) [`8eaf410`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F8eaf41069264753947e5c9633b567e589dc0e532) 感谢 [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - 修复 (SDK)：对于零元 onFinish 回调，跳过 post-stream 的 getHistory 调用。\n\n-   更新了依赖项 \\[[`8eaf410`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F8eaf41069264753947e5c9633b567e589dc0e532)]:\n    -   @langchain\u002Flanggraph-sdk@1.8.2\n","2026-03-28T02:56:27",{"id":192,"version":193,"summary_zh":194,"released_at":195},53445,"@langchain\u002Flanggraph-sdk@1.8.2","### 补丁变更\n\n-   [#2250](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2250) [`8eaf410`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F8eaf41069264753947e5c9633b567e589dc0e532) 感谢 [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - 修复（SDK）：对于零元 onFinish 回调，跳过 post-stream 的 getHistory 调用\n","2026-03-28T02:56:24",{"id":197,"version":198,"summary_zh":199,"released_at":200},53446,"@langchain\u002Flanggraph-cli@1.1.17","### 补丁变更\n\n-   [#2247](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2247) [`9874420`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F9874420019199a7064501b53b9407bd23dc752f9) 感谢 [@jdrogers940](https:\u002F\u002Fgithub.com\u002Fjdrogers940)! - 尊重已设置的 HTTP 配置。同时，当 .env 文件缺失时不再记录错误。\n\n-   更新了依赖项 \\[]:\n    -   @langchain\u002Flanggraph-api@1.1.17\n","2026-03-27T17:57:46",{"id":202,"version":203,"summary_zh":204,"released_at":205},53447,"@langchain\u002Flanggraph-api@1.1.17","### 补丁变更\n\n-   更新了依赖项 \\[]：\n    -   @langchain\u002Flanggraph-ui@1.1.17\n","2026-03-27T17:57:43",{"id":207,"version":208,"summary_zh":209,"released_at":210},53448,"@langchain\u002Flanggraph-ui@1.1.17","\n","2026-03-27T17:57:40",{"id":212,"version":213,"summary_zh":214,"released_at":215},53449,"@langchain\u002Flanggraph-sdk@1.8.1","### 补丁变更\n\n-   [#2237](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2237) [`88726df`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F88726dfe222aed64e5cd5dfa6f77f886b5a0d205) 感谢 [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - 将共享的 `WithClassMessages\u003CT>` 类型提取到 `@langchain\u002Flanggraph-sdk\u002Fui` 中\n\n-   [#2243](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2243) [`7dfcbff`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F7dfcbffd4805b2b4cc41f07f30be57ed732786b4) 感谢 [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - 修复 (sdk): 规范化 JavaScript 和 Python 之间的中断处理逻辑","2026-03-26T17:41:15",{"id":217,"version":218,"summary_zh":219,"released_at":220},53450,"@langchain\u002Fsvelte@0.3.1","### 补丁变更\n\n-   [#2237](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2237) [`88726df`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F88726dfe222aed64e5cd5dfa6f77f886b5a0d205) 感谢 [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - 将共享的 `WithClassMessages\u003CT>` 类型提取到 `@langchain\u002Flanggraph-sdk\u002Fui` 中\n\n-   更新了依赖项 \\[[`88726df`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F88726dfe222aed64e5cd5dfa6f77f886b5a0d205), [`7dfcbff`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F7dfcbffd4805b2b4cc41f07f30be57ed732786b4)]:\n    -   @langchain\u002Flanggraph-sdk@1.8.1\n","2026-03-26T17:41:12",{"id":222,"version":223,"summary_zh":224,"released_at":225},53451,"@langchain\u002Fangular@0.3.1","### Patch Changes\n\n-   [#2237](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2237) [`88726df`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F88726dfe222aed64e5cd5dfa6f77f886b5a0d205) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - Extract shared `WithClassMessages\u003CT>` type to `@langchain\u002Flanggraph-sdk\u002Fui`\n\n-   [#2235](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2235) [`1b597f1`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F1b597f124e8557cef86f98670ca594237e0040cf) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - fix(@langchain\u002Fangular): resolve non-idiomatic patterns\n\n-   [#2234](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2234) [`c52ba27`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002Fc52ba278ebcaab7ee098e4afae00d120aca96ef7) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - fix(angular): sync isLoading signal in custom transport\n\n-   Updated dependencies \\[[`88726df`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F88726dfe222aed64e5cd5dfa6f77f886b5a0d205), [`7dfcbff`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F7dfcbffd4805b2b4cc41f07f30be57ed732786b4)]:\n    -   @langchain\u002Flanggraph-sdk@1.8.1\n","2026-03-26T17:41:09",{"id":227,"version":228,"summary_zh":229,"released_at":230},53452,"@langchain\u002Fvue@0.3.1","### Patch Changes\n\n-   [#2237](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2237) [`88726df`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F88726dfe222aed64e5cd5dfa6f77f886b5a0d205) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - Extract shared `WithClassMessages\u003CT>` type to `@langchain\u002Flanggraph-sdk\u002Fui`\n\n-   [#2236](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2236) [`4a80eef`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F4a80eefa4463b2043b50892b1ad460b6aa0a168c) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - fix(vue): use computed() and reactive refs in custom transport\n\n-   Updated dependencies \\[[`88726df`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F88726dfe222aed64e5cd5dfa6f77f886b5a0d205), [`7dfcbff`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F7dfcbffd4805b2b4cc41f07f30be57ed732786b4)]:\n    -   @langchain\u002Flanggraph-sdk@1.8.1\n","2026-03-26T17:41:06",{"id":232,"version":233,"summary_zh":234,"released_at":235},53453,"@langchain\u002Flanggraph@1.2.6","### Patch Changes\n\n-   [#2241](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2241) [`6ee23e8`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F6ee23e819b5da43a5a0c62560f85a9037a427630) Thanks [@pawel-twardziak](https:\u002F\u002Fgithub.com\u002Fpawel-twardziak)! - feat: add browser support for interrupt, writer, and other Node-only exports\n\n    Export `interrupt`, `writer`, `pushMessage`, `getStore`, `getWriter`, `getConfig`, `getPreviousState`, `getCurrentTaskInput` from `web.ts` and add a `\"browser\"` condition to the `\".\"` package export so browser bundlers resolve to `web.js` instead of pulling in `node:async_hooks`.\n\n-   [#2245](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2245) [`77af976`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F77af97650c0f1671338911994c2e355b29489528) Thanks [@hntrl](https:\u002F\u002Fgithub.com\u002Fhntrl)! - revert abort signal change that was causing problematic errors\n\n-   [#2242](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2242) [`bdcf290`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002Fbdcf290198ce5cea4367ee8c9f1cbbbcf14d05e4) Thanks [@hntrl](https:\u002F\u002Fgithub.com\u002Fhntrl)! - clean up resolved checkpointer promises to reduce memory retention\n\n-   Updated dependencies \\[[`88726df`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F88726dfe222aed64e5cd5dfa6f77f886b5a0d205), [`7dfcbff`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F7dfcbffd4805b2b4cc41f07f30be57ed732786b4)]:\n    -   @langchain\u002Flanggraph-sdk@1.8.1\n","2026-03-26T17:41:03",{"id":237,"version":238,"summary_zh":239,"released_at":240},53454,"@langchain\u002Freact@0.2.2","### Patch Changes\n\n-   [#2237](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2237) [`88726df`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F88726dfe222aed64e5cd5dfa6f77f886b5a0d205) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - Extract shared `WithClassMessages\u003CT>` type to `@langchain\u002Flanggraph-sdk\u002Fui`\n\n-   [#2243](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2243) [`7dfcbff`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F7dfcbffd4805b2b4cc41f07f30be57ed732786b4) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - fix(skd): normalize interrupts between JS and Python\n\n-   Updated dependencies \\[[`88726df`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F88726dfe222aed64e5cd5dfa6f77f886b5a0d205), [`7dfcbff`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F7dfcbffd4805b2b4cc41f07f30be57ed732786b4)]:\n    -   @langchain\u002Flanggraph-sdk@1.8.1\n","2026-03-26T17:41:00",{"id":242,"version":243,"summary_zh":244,"released_at":245},53455,"@langchain\u002Flanggraph-sdk@1.8.0","### Minor Changes\n\n-   [#2227](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2227) [`414a7ad`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F414a7adf908ba4f7ffef4985df3a95f14202591b) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - feat: extract shared orchestrator to eliminate duplicated code across SDK packages\n","2026-03-20T22:28:57",{"id":247,"version":248,"summary_zh":249,"released_at":250},53456,"@langchain\u002Freact@0.2.1","### Patch Changes\n\n-   [#2228](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2228) [`07e9044`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F07e9044487aeed6f6b40b2b49a52615cda90dcc1) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - Fix @langchain\u002Freact hook dispatch and Suspense cache ergonomics\n\n-   Updated dependencies \\[[`414a7ad`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F414a7adf908ba4f7ffef4985df3a95f14202591b)]:\n    -   @langchain\u002Flanggraph-sdk@1.8.0\n","2026-03-20T22:28:54",{"id":252,"version":253,"summary_zh":254,"released_at":255},53457,"@langchain\u002Fvue@0.3.0","### Minor Changes\n\n-   [#2227](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2227) [`414a7ad`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F414a7adf908ba4f7ffef4985df3a95f14202591b) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - feat: extract shared orchestrator to eliminate duplicated code across SDK packages\n\n### Patch Changes\n\n-   [#2230](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2230) [`11acfc1`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F11acfc137b65c7e28c3fab61444c00db3e4be229) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - fix(vue): integrate LangChainPlugin with useStream and fix subagent reactivity\n\n-   Updated dependencies \\[[`414a7ad`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F414a7adf908ba4f7ffef4985df3a95f14202591b)]:\n    -   @langchain\u002Flanggraph-sdk@1.8.0\n","2026-03-20T22:28:51",{"id":257,"version":258,"summary_zh":259,"released_at":260},53458,"@langchain\u002Fangular@0.3.0","### Minor Changes\n\n-   [#2227](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2227) [`414a7ad`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F414a7adf908ba4f7ffef4985df3a95f14202591b) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - feat: extract shared orchestrator to eliminate duplicated code across SDK packages\n\n-   [#2226](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2226) [`13d117e`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F13d117ec90f6cf2f5d11584d298d32bb3ff160e1) Thanks [@christian-bromann](https:\u002F\u002Fgithub.com\u002Fchristian-bromann)! - feat(angular): use `injectStream`\u002F`injectStreamCustom`\n\n### Patch Changes\n\n-   Updated dependencies \\[[`414a7ad`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F414a7adf908ba4f7ffef4985df3a95f14202591b)]:\n    -   @langchain\u002Flanggraph-sdk@1.8.0\n","2026-03-20T22:28:48",{"id":262,"version":263,"summary_zh":264,"released_at":265},53459,"@langchain\u002Flanggraph@1.2.5","### Patch Changes\n\n-   [#2213](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2213) [`a09932a`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002Fa09932a203062d52e98e6dc5fd80ab572b123700) Thanks [@hntrl](https:\u002F\u002Fgithub.com\u002Fhntrl)! - fix(core): prevent AbortSignal listener leak in stream() and streamEvents()\n\n    `Pregel.stream()` and `streamEvents()` called `combineAbortSignals()` but discarded the `dispose` function, leaking one abort listener on the caller's signal per invocation. Over many invocations this caused unbounded memory growth as each leaked listener retained references to its associated graph execution state.\n\n    -   Use `AbortSignal.any()` on Node 20+ which handles listener lifecycle automatically via GC\n    -   Fall back to manual listener management on Node 18, with proper `dispose()` called when the stream completes or is cancelled\n\n-   [#2210](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fpull\u002F2210) [`4d2e948`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F4d2e9483208e105b7c45ab1cbc8ac8d540fbb23d) Thanks [@jackjin1997](https:\u002F\u002Fgithub.com\u002Fjackjin1997)! - Fix `AnyValue.update()` returning `false` instead of `true` when values are received, aligning with all other channel implementations.\n\n-   Updated dependencies \\[[`414a7ad`](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraphjs\u002Fcommit\u002F414a7adf908ba4f7ffef4985df3a95f14202591b)]:\n    -   @langchain\u002Flanggraph-sdk@1.8.0\n","2026-03-20T22:28:45"]