openui-forge-vercel
installsforks
How to install openui-forge-vercel?
othmanadi/openui-forge/openui-forge-vercel
$npx -y skills add OthmanAdi/openui-forge --skill openui-forge-vercelOpenUI generative UI with Vercel AI SDK. streamText, toUIMessageStreamResponse, and tools support.
SKILL.md
| 1 | # OpenUI Forge — Vercel AI SDK |
| 2 | |
| 3 | Build generative UI apps with OpenUI + Vercel AI SDK. Native streaming with `streamText` and `toUIMessageStreamResponse()`. |
| 4 | |
| 5 | ## Activation Triggers |
| 6 | |
| 7 | - "openui vercel", "openui vercel ai", "openui ai sdk" |
| 8 | - "generative ui vercel", "vercel ai streaming ui" |
| 9 | - "useChat openui", "streamText openui" |
| 10 | |
| 11 | ## Prerequisites |
| 12 | |
| 13 | - Node.js >= 22 (24 LTS recommended), React >= 18.3.1 (19+ recommended) |
| 14 | - `OPENAI_API_KEY` environment variable set |
| 15 | - Next.js project (App Router) |
| 16 | |
| 17 | ## Quick Start |
| 18 | |
| 19 | 1. Install dependencies: |
| 20 | ```bash |
| 21 | npm install @openuidev/react-ui @openuidev/react-lang lucide-react zod ai @ai-sdk/openai @ai-sdk/react |
| 22 | ``` |
| 23 | Pin to the AI SDK v6 line: `ai@^6`, `@ai-sdk/openai@^3`, `@ai-sdk/react@^3`. |
| 24 | 2. Add the CSS import to `app/layout.tsx`: |
| 25 | ```tsx |
| 26 | import "@openuidev/react-ui/components.css"; |
| 27 | ``` |
| 28 | 3. Create the API route and frontend page below |
| 29 | 4. Run `npm run dev` and test |
| 30 | |
| 31 | ## Full Code |
| 32 | |
| 33 | ### Backend: `app/api/chat/route.ts` |
| 34 | |
| 35 | ```typescript |
| 36 | import { openuiChatLibrary } from "@openuidev/react-ui/genui-lib"; |
| 37 | import { convertToModelMessages, streamText } from "ai"; |
| 38 | import { openai } from "@ai-sdk/openai"; |
| 39 | |
| 40 | export async function POST(req: Request) { |
| 41 | const { messages } = await req.json(); |
| 42 | |
| 43 | const systemPrompt = openuiChatLibrary.prompt({ |
| 44 | preamble: "You are a helpful assistant that generates interactive UIs.", |
| 45 | additionalRules: ["Always use Stack as root when combining multiple components."], |
| 46 | }); |
| 47 | |
| 48 | // AI SDK v6: convert the UI message stream into model messages before passing to the model. |
| 49 | const modelMessages = await convertToModelMessages(messages); |
| 50 | |
| 51 | const result = streamText({ |
| 52 | model: openai(process.env.OPENAI_MODEL ?? "gpt-5.5"), |
| 53 | system: systemPrompt, |
| 54 | messages: modelMessages, |
| 55 | }); |
| 56 | |
| 57 | return result.toUIMessageStreamResponse(); |
| 58 | } |
| 59 | ``` |
| 60 | |
| 61 | ### Backend with Tools: `app/api/chat/route.ts` |
| 62 | |
| 63 | ```typescript |
| 64 | import { openuiChatLibrary } from "@openuidev/react-ui/genui-lib"; |
| 65 | import { convertToModelMessages, streamText, tool, stepCountIs } from "ai"; |
| 66 | import { openai } from "@ai-sdk/openai"; |
| 67 | import { z } from "zod"; |
| 68 | |
| 69 | export async function POST(req: Request) { |
| 70 | const { messages } = await req.json(); |
| 71 | |
| 72 | const systemPrompt = openuiChatLibrary.prompt({ |
| 73 | preamble: "You are a helpful assistant that generates interactive UIs. Use tools to fetch data before rendering.", |
| 74 | }); |
| 75 | |
| 76 | // AI SDK v6: convert the UI message stream into model messages before passing to the model. |
| 77 | const modelMessages = await convertToModelMessages(messages); |
| 78 | |
| 79 | const result = streamText({ |
| 80 | model: openai(process.env.OPENAI_MODEL ?? "gpt-5.5"), |
| 81 | system: systemPrompt, |
| 82 | messages: modelMessages, |
| 83 | tools: { |
| 84 | getWeather: tool({ |
| 85 | description: "Get current weather for a city", |
| 86 | inputSchema: z.object({ |
| 87 | city: z.string().describe("City name"), |
| 88 | }), |
| 89 | execute: async ({ city }) => { |
| 90 | return { city, temp: 22, condition: "sunny" }; |
| 91 | }, |
| 92 | }), |
| 93 | }, |
| 94 | // AI SDK v6: stopWhen replaces the removed `maxSteps` option. |
| 95 | stopWhen: stepCountIs(3), |
| 96 | }); |
| 97 | |
| 98 | return result.toUIMessageStreamResponse(); |
| 99 | } |
| 100 | ``` |
| 101 | |
| 102 | ### Frontend (useChat + Renderer): `app/chat/page.tsx` |
| 103 | |
| 104 | Drive the conversation with `useChat` from `@ai-sdk/react`, then render each assistant message with a per-message `<Renderer>` from `@openuidev/react-lang`. The Renderer takes the assistant text as `response`, the component library as `library` (NOT `componentLibrary`), an `isStreaming` flag for the in-flight message, and an `onAction` handler for built-in actions like continuing the conversation. |
| 105 | |
| 106 | ```tsx |
| 107 | "use client"; |
| 108 | import { useChat } from "@ai-sdk/react"; |
| 109 | import { Renderer, BuiltinActionType } from "@openuidev/react-lang"; |
| 110 | import type { ActionEvent } from "@openuidev/react-lang"; |
| 111 | import { openuiChatLibrary } from "@openuidev/react-ui/genui-lib"; |
| 112 | import { useState } from "react"; |
| 113 | |
| 114 | export default function ChatPage() { |
| 115 | const [input, setInput] = useState(""); |
| 116 | const { messages, sendMessage, status } = useChat(); |
| 117 | const isLoading = status === "submitted" || status === "streaming"; |
| 118 | |
| 119 | const handleSend = (text: string) => { |
| 120 | const trimmed = text.trim(); |
| 121 | if (!trimmed || isLoading) return; |
| 122 | setInput(""); |
| 123 | sendMessage({ text: trimmed }); |
| 124 | }; |
| 125 | |
| 126 | const handleAction = (event: ActionEvent) => { |
| 127 | if (event.type === BuiltinActionType.ContinueConversation && event.humanFriendlyMessage) { |
| 128 | handleSend(event.humanFriendlyMessage); |
| 129 | } |
| 130 | }; |
| 131 | |
| 132 | return ( |
| 133 | <div> |
| 134 | {messages.map((message, i) => { |
| 135 | const isLast = i === messages.length - 1; |
| 136 | |
| 137 | if (message.role === "user") { |
| 138 | const text = message.parts |
| 139 | .filter((p): p is { type: "text"; text: string } => p.type === "text") |
| 140 | .map((p) => p.text) |
| 141 | .join(""); |
| 142 | return <div key={message.id}>{text}</div>; |
People who liked this also viewed
6 picksType
- web-design-guidelinesReview UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check accessibility", "audit design", "review UX", or "check my site against best practices".SkillsJul 2026
- lark-slides飞书幻灯片：创建和编辑幻灯片。创建演示文稿、读取幻灯片内容、管理幻灯片页面（创建、删除、读取、局部替换）。当用户需要创建或编辑幻灯片、读取或修改单个页面时使用。当用户给出 doubao.com 的 /slides/ URL/token 时，也应直接使用本 skill，不要因为域名不是飞书而回退到 WebFetch；路由依据是 URL 路径模式和 token，而不是域名。不负责：云文档内容编辑（走 SkillsJul 2026
- design-taste-frontendTaste-Skill - gives your AI good taste. stops the AI from generating boring, generic slop SkillsJul 2026
- ui-ux-pro-maxUI/UX design intelligence for web and mobile. Searchable local database with 84 styles, 192 color palettes, 74 font pairings, 192 product types, 98 UX guidelines, 104 icon entries, 16 GSAP motion presSkillsJul 2026
- caveman-helpQuick-reference card for all caveman modes, skills, and commands. One-shot display, not a persistent mode. Trigger: /caveman-help, "caveman help", "what caveman commands", "how do I use caveman".SkillsJul 2026
- codebase-designShared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-nSkillsJul 2026
