--- title: "OpenAI DevDay unveils major new product: Agent Builder is here! Build AI applications with drag-and-drop" type: "News" locale: "zh-CN" url: "https://longbridge.com/zh-CN/news/260032524.md" description: "OpenAI will launch a new tool called Agent Builder on October 6th at DevDay, allowing users to build AI agent workflows through a visual canvas with drag-and-drop functionality. Users can connect models, external tools, and ChatKit widgets to orchestrate complex AI tasks. Agent Builder offers various preset templates and modules, supporting logical nodes, connectors, and security tools, allowing users to debug in preview mode before publishing. The strength of this tool lies in its integration capabilities with third-party applications and services" datetime: "2025-10-06T09:35:38.000Z" locales: - [zh-CN](https://longbridge.com/zh-CN/news/260032524.md) - [en](https://longbridge.com/en/news/260032524.md) - [zh-HK](https://longbridge.com/zh-HK/news/260032524.md) --- > 支持的语言: [English](https://longbridge.com/en/news/260032524.md) | [繁體中文](https://longbridge.com/zh-HK/news/260032524.md) # OpenAI DevDay unveils major new product: Agent Builder is here! Build AI applications with drag-and-drop OpenAI is set to launch a new tool called **Agent Builder** on October 6 (North American time) at OpenAI DevDay. According to the latest leaks, this tool will allow users to build their own AI agents' workflows through a visual canvas using a drag-and-drop method. Users will be able to connect models, external tools, ChatKit widgets, etc., to orchestrate complex AI tasks in a building block manner. From the exposed interface, its direct competitors will be automation platforms like n8n and Zapier. ## Drag-and-Drop Canvas for Visual AI Agent Construction The core of Agent Builder is a drag-and-drop canvas interface, where users can start creating their workflows from predefined templates and modules. According to leaked screenshots, Agent Builder offers various preset templates, including: **Customer Service**: Resolve customer inquiries using custom strategies. **Data Augmentation**: Integrate data to answer user questions. **Structured Data Q&A**: Query databases using natural language. **Document Comparison**: Analyze and highlight differences between uploaded documents. On the canvas, users can freely combine various modular building blocks, including: **Logic Nodes**: Such as if-else conditional judgments, loops, etc. **Connectors**: Such as MCP, used to connect external services. **Tool Nodes**: Such as user approval, file search, data transformation, etc. **Security Tools**: Built-in modules like Guardrails to ensure the output quality of the Agent. The entire interface is very intuitive, allowing users to drag the required modules from the sidebar component library and connect them like a flowchart to define a complete Agent workflow. In terms of deployment, users can first debug in preview and testing mode before publishing for broader use. ## Connecting Everything to Build an AI Ecosystem The power of Agent Builder lies not only in its visual workflow orchestration but also in its strong connectivity capabilities. Through MCP, Agent Builder can easily integrate with various third-party applications and services. The exposed interface shows that OpenAI officially provides connectors for Gmail, Google Calendar, Google Drive, Outlook, SharePoint, etc. It also supports third-party services created and maintained by other developers This means that developers can easily integrate OpenAI's model capabilities with their existing work software and data sources to build automated agents that can perform real tasks. ## Goal: Lower the Barriers to AI Application Development The launch of Agent Builder is seen as a significant expansion of OpenAI's platform capabilities. It marks OpenAI's transition from being a pure model API provider to a broader AI ecosystem builder. The most direct beneficiaries of this tool will be developers, solution architects, and enterprises already using the OpenAI API. It offers a more intuitive and efficient alternative to manually coding integrations. Its core goal is to lower the barriers to creating production-grade AI agents, allowing non-experts to quickly prototype and get started, thereby accelerating the iteration, testing, and operation of AI-driven solutions. In an already crowded automation market, OpenAI's bet is to establish a unique competitive advantage among new and existing customers through deep integration of its own AI models, excellent ease of use, and pre-set guardrails and other logical modules. All suspense will be revealed at tomorrow's DevDay. Author of this article: AI Cambrian, Source: AI Cambrian, Original Title: "OpenAI DevDay Major New Product Revealed: Agent Builder is Here! Drag and Drop to Build AI Applications" Risk Warning and Disclaimer The market has risks, and investment requires caution. This article does not constitute personal investment advice and does not take into account individual users' specific investment goals, financial situations, or needs. Users should consider whether any opinions, views, or conclusions in this article align with their specific circumstances. Investing based on this is at your own risk ### 相关股票 - [OpenAI (OpenAI.NA)](https://longbridge.com/zh-CN/quote/OpenAI.NA.md) ## 相关资讯与研究 - [Anthropic officially bans OpenClaw, causing a global developer crisis within 24 hours.](https://longbridge.com/zh-CN/news/281674328.md) - [Anthropic eyes its own version of OpenClaw - Semafor](https://longbridge.com/zh-CN/news/281661712.md) - [OpenAI, not yet public, raises $3B from retail investors in monster $122B fund raise](https://longbridge.com/zh-CN/news/281250634.md) - [OpenAI is now bringing in $2 billion a month - and 3 more highlights from its latest update](https://longbridge.com/zh-CN/news/281258023.md) - [Checkout.com Pushes AI Payment Standards as Chinese Sellers Expand Overseas](https://longbridge.com/zh-CN/news/281354478.md)