AI聊天UI渲染服务

BranderUX提供的AI聊天UI渲染服务,支持在Claude、ChatGPT等MCP兼容的聊天平台上直接渲染品牌化的交互式UI组件。

  • by BranderUX
  • TypeScript
  • 本地部署
  • 更新于
在 GitHub 查看 BranderUX/mcp-demo

使用说明

@brander/mcp-demo

A demo MCP server showcasing BranderUX — branded, interactive UI rendering inside AI chatbots. Uses @brander/mcp-tools to render rich UI components directly in conversations on Claude, ChatGPT, and other MCP-compatible hosts.

Connect

Claude.ai / ChatGPT (Remote)

Use the hosted URL — no installation needed:

https://mcp-demo.branderux.com/mcp

Claude Desktop (Local)

Add to your config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "branderux-demo": {
      "command": "npx",
      "args": ["-y", "@brander/mcp-demo"],
      "env": {
        "BRANDER_PROJECT_ID": "your_project_id",
        "BRANDER_API_KEY": "bux_pk_your_key"
      }
    }
  }
}

Tools

Tool Description
generate_screen Render branded UI screens with 15 element types (charts, tables, grids, forms, cards, etc.)
learn_about_branderux Learn about BranderUX — platform overview, integration methods, UI generation modes, AI features, no-code philosophy
explore_elements Browse all 15 element types with descriptions, use cases, and sample data ready for generate_screen
get_demo_scenario Get pre-built multi-element screen compositions (dashboards, catalogs, reports, showcases)
get_integration_guide Step-by-step integration guides with real code examples for SDK, MCP Tools, and Embed methods

Try It

Connect to the demo and try these prompts:

  • "What is BranderUX?"
  • "Show me all available elements"
  • "Show me a sales dashboard"
  • "Create a product catalog with items and pricing"
  • "How do I integrate BranderUX into my app?"
  • "Show me an analytics report with charts"

Demo Scenarios

Scenario Description
sales-dashboard Revenue metrics, trend charts, and transaction tables
product-catalog Item grid with categories and pricing
analytics-report Multi-chart report with KPIs
branderux-overview BranderUX capabilities rendered in its own elements
element-showcase All 15 element types displayed as a product catalog
integration-comparison SDK vs MCP vs Embed side-by-side comparison

How It Works

BranderUX uses the MCP Apps extension to render interactive UI inside AI conversations. When the AI calls generate_screen, it returns:

  1. Text summary — for the AI to understand what was rendered
  2. Structured content — element data (charts, tables, forms, etc.)
  3. HTML resource — a bundled React app that renders the branded UI in a sandboxed iframe

All UI is styled with your project's brand colors, fonts, and layout — configured in the BranderUX dashboard. No developers needed — AI agents handle the design, PMs control everything through the UI.

Integration Pattern

Adding BranderUX to any MCP server is one line:

import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
import { registerBranderTools } from "@brander/mcp-tools";

const server = new McpServer({ name: "my-server", version: "1.0.0" });

// Add your own tools...

// One line — branded UI rendering for 15 element types
await registerBranderTools(server, {
  projectId: process.env.BRANDER_PROJECT_ID!,
  apiKey: process.env.BRANDER_API_KEY!,
});

Environment Variables

Variable Required Description
BRANDER_PROJECT_ID Yes Your BranderUX project ID
BRANDER_API_KEY Yes Project API key (bux_pk_...) — generate under Projects → API Keys. Legacy BRANDER_BETA_KEY (bux_dp_...) still works as a fallback.

License

MIT