AI
AI SDK Patterns

Command Palette

Search for a command to run...

PatternsComposeGitHub
All Patterns

Core / SDK

9
  • Structured Output
  • Generative UI
  • JSON Renderer
  • Text Generation
  • Image Generation
  • Streaming Object
  • Code Artifact
  • Form Generator
  • CSV Editor

Chat

4
  • Streaming Chat
  • Markdown Chat
  • Reasoning Display
  • Chat with Citations

Agents

5
  • Tool Calling
  • Multi-Step Agent
  • Routing Agent
  • Orchestrator Agent
  • Evaluator-Optimizer

Tools

2
  • Web Search Agent
  • RAG Pipeline

Workflows

3
  • Human-in-the-Loop
  • Sequential Workflow
  • Parallel Workflow
PatternsComposeAI SDK DocsGitHub

Built with AI SDK · shadcn/ui · Next.js

© 2026 AI SDK Patterns

RAG Pipeline

Retrieval-Augmented Generation: embed documents, store in a vector database, and retrieve context for grounded AI responses.

toolsadvancedragembeddingsvector-searchretrieval

Loading interactive preview...

Installation

Option 1: Install via CLI

pnpm dlx shadcn@latest add https://ai-sdk-patterns.vercel.app/r/rag-pipeline

Automatically installs the pattern and its dependencies in your project.

Option 2: Copy or Download

Download the complete pattern as a standalone Next.js project.

Usage

1. Set up environment variables

# .env.local
ANTHROPIC_API_KEY=your_anthropic_key
OPENAI_API_KEY=your_openai_key
GOOGLE_GENERATIVE_AI_API_KEY=your_google_key

Add your AI provider API key to enable real functionality.

2. Run the development server

npm run dev
# or
pnpm dev
# or
yarn dev

Open http://localhost:3000 to see the pattern in action.

3. Customize for your needs

The pattern is ready to use. Modify the components, API routes, and styling to fit your application.

  • Update the UI components in app/page.tsx
  • Modify API logic in app/api/
  • Adjust styling with Tailwind CSS classes
  • Add your own business logic and data sources

Use Cases

Document Q&A Systems

Build question-answering systems that search through company documents, legal papers, or technical manuals for accurate responses.

Knowledge Base Assistants

Create support systems that retrieve relevant information from documentation, FAQs, and help articles to answer user queries.

Research Paper Analysis

Develop tools that search academic papers, extract relevant passages, and provide context-aware answers to research questions.

Technical Documentation Helpers

Build assistants that search API documentation, code repositories, and technical guides to provide accurate coding help.

Technical Details

Dependencies

• Next.js 16+ (App Router)
• AI SDK v6
• React 18+
• Tailwind CSS
• TypeScript

Files Included

• app/page.tsx
• app/api/rag/route.ts
• app/api/rag/index/route.ts
• lib/vector-store.ts

Related patterns

Web Search Agent

Integrate third-party search APIs (Tavily, Exa) with the AI SDK to build a grounded search agent with source citations.

Tool Calling

Let the AI model call functions and use external tools with the multi-step tool loop.

Interactive Demo Coming Soon

The rag-pipeline pattern demo is being prepared.

Download the pattern to try it with your API key