\n A powerful SDK for Model Context Protocol (MCP) by Aintent.ai, an Axes Labs company.\n
\n\n\n Features •\n Quick Start •\n Installation •\n Usage •\n Widgets •\n API •\n Examples •\n Documentation\n
\n\n## Features ✨\n\n- 🎯 **Intent Processing**: Advanced natural language understanding and intent classification\n- 🔄 **Workflow Generation**: Automated workflow creation from natural language inputs\n- 🎨 **Interactive Widgets**: Ready-to-use UI components for chat and flow visualization\n- 🛠️ **Extensible Architecture**: Easy to customize and extend with your own components\n- 📦 **TypeScript Ready**: Full TypeScript support with comprehensive type definitions\n- 🔌 **Plugin System**: Extensible plugin architecture for custom integrations\n- 🎭 **Theme Support**: Light and dark themes out of the box\n- 📱 **Responsive Design**: Works seamlessly on desktop and mobile\n\n## Quick Start 🚀\n\n```typescript\nimport { MCPClient, ChatWidget } from '@aintent/mcp-kit';\n\n// Initialize the client\nconst client = new MCPClient({\n apiKey: 'your-api-key'\n});\n\n// Create a chat widget\nconst chat = new ChatWidget('#chat-container', client, {\n theme: 'light',\n height: '500px'\n});\n\n// Process intents and generate workflows\nchat.on('message', async (message) => {\n const intent = await client.processIntent(message);\n const flow = await client.generateWorkflow(intent);\n await client.executeWorkflow(flow);\n});\n```\n\n## Installation 📦\n\n```bash\n# Using npm\nnpm install @aintent/mcp-kit\n\n# Using yarn\nyarn add @aintent/mcp-kit\n\n# Using pnpm\npnpm add @aintent/mcp-kit\n```\n\n## Usage 💡\n\n### Core Client\n\n```typescript\nimport { MCPClient } from '@aintent/mcp-kit';\n\nconst client = new MCPClient({\n apiKey: 'your-api-key',\n options: {\n timeout: 5000,\n retries: 3\n }\n});\n\n// Process an intent\nconst intent = await client.processIntent('Deploy the application to production');\n\n// Generate a workflow\nconst flow = await client.generateWorkflow(intent);\n\n// Execute the workflow\nconst result = await client.executeWorkflow(flow);\n```\n\n### Interactive Chat\n\n```typescript\nimport { ChatWidget } from '@aintent/mcp-kit';\n\nconst chat = new ChatWidget('#chat', client, {\n theme: 'dark',\n height: '600px',\n placeholder: 'What would you like to do?',\n maxMessages: 100,\n autoScroll: true\n});\n\n// Listen for events\nchat.on('message', (message) => {\n console.log('New message:', message);\n});\n```\n\n### Flow Visualization\n\n```typescript\nimport { FlowViewerWidget } from '@aintent/mcp-kit';\n\nconst viewer = new FlowViewerWidget('#viewer', flow, {\n height: '800px',\n theme: 'light',\n zoom: {\n min: 0.5,\n max: 3,\n step: 0.1\n }\n});\n\n// Update flow data\nviewer.updateFlow(newFlow);\n```\n\n## Widgets 🎨\n\nThe SDK includes ready-to-use UI components:\n\n- **ChatWidget**: Interactive chat interface for natural language interactions\n- **FlowViewerWidget**: Visual workflow representation with zoom and pan capabilities\n- **Custom Widgets**: Extend `MCPWidget` to create your own components\n\n[Learn more about widgets →](./docs/widgets/README.md)\n\n## API Reference 📚\n\n### MCPClient\n\n- `processIntent(input: string): PromiseMade with ❤️ by Aintent.ai, an Axes Labs company
\n\n\n Website •\n Documentation •\n GitHub\n
\n","readmeFilename":"README.md","_rev":"1-8e973f1a050ab208b8de16fff71f0864"}