Did OpenAI Just End N8N?!

New workflow Draft Core Agent End Note Tools File search Guardrails MCP Logic If / else While User approval Data Transform Set state Simple workflow to research stuff on web search and provide summary. Example input: "Analyze Tesla" Start Loop while a condition is true Did OpenAI Just End N8N?! did openai just kill n8n and make? October 12, 2023 Introducing AgentKit New tools for building, deploying, and optimizing agents. Listen to article 5:23 Today we're launching AgentKit, a complete set of tools for developers and enterprises to build, deploy, and optimize multi-step AI agents. This includes a suite of fragmented tools - simple, production-ready, versioning, custom connectors, remote execution, prompt tuning, and weeks of frontier models. With AgentKit, developers can now design workflows visually and embed scientific, LLM-faster models like: Ask ChatGPT OpenAI Research Safety For Business For Developers ChatGPT Store News Introducing AgentKit New tools for building, deploying, and optimizing agents. We're also expanding evaluation capabilities with new features like controls, more grading, assisted prompt optimization, and fine-tuning controls to support its precision and improve agent performance. Since releasing the Response API and Agents SDK in December, we've seen incredible innovation in how developers build end-to-end workflows for deep research, customer support, and more. These have built a "pro-agent" that can be used in a "chat agent" workflow. AgentKit builds on the Response API and Agents SDK to provide a more cohesive, flexible, and widely Design workflows with Agent Builder In agent workflows grow more complex, developers said need a visual canvas to build workflows. They can drag and drop nodes to connect agents, tools, and control-flow logic. When you're ready to deploy, embed the workflow into your site with ChatGPT, or download the SDK code to run it yourself. inside of chatgpt. but what does mean for n8n and make users? Apify Run AI Agents API Lead Generation API It seems like agentkit is just another drag-and-drop, no code agent builder. It uses nodes, logic and guardrails. However, you can use widgets inside your agent response, which is a useful feature. but if you're n8n or another automation platform, I wouldn't be cancelling my subscriptions just yet. Structured output (JSON) This model will generate a JSON object that matches this schema. Name response_schema Properties Name Type Description classification STR Add description + Add Generate agentkit is still in beta, so I'd definitely be keeping an add new features. platform.openai.com/docs/overview Personal Default project javascript import OpenAI from "openai"; const client = new OpenAI(); async function main() { const response = await client.chat.completions.create({ model: "gpt-5", input: "Write a short bedtime story about a unicorn.", }); console.log(response.choices[0].message.content); } main(); Build with AgentKit Design and version workflows with agent builder or use the SDK for code-level control. + More Browse Models platform.openai.com/docs/guides/agent-builder Agent Builder Build a chat agent workflow with custom logic and tools + Create AI Assistant AI Assistant AI Assistant Customer Service AI Assistant AI Assistant AI Assistant Structured Data Q&A Document companion Internal knowledge resort Templates Or you can create your own agent. click on this. Open Agent Builder There are three main steps in building agents: 1. Design a workflow in Agent Builder. This is where you work. 2. Publish your workflow. It's an object with a unique ID you can deploy when you're ready. 3. Deploy your workflow. Pass the ID into your site or download the Agents SDK code to run it yourself.