AI SDK Examples - Supermemory

This page provides comprehensive examples of using Supermemory with the Vercel AI SDK, covering Memory Tools and User Profiles approaches.

Personal Assistant with Memory Tools

Build an AI assistant that remembers user preferences and past interactions:

typescript

import { streamText } from 'ai'
import { createAnthropic } from '@ai-sdk/anthropic'
import { supermemoryTools } from '@supermemory/tools/ai-sdk'

const anthropic = createAnthropic({
  apiKey: process.env.ANTHROPIC_API_KEY!
})

export async function POST(request: Request) {
  const { messages } = await request.json()

  const result = await streamText({
    model: anthropic('claude-3-sonnet-20240229'),
    messages,
    tools: supermemoryTools(process.env.SUPERMEMORY_API_KEY!),
    system: `You are a helpful personal assistant. When users share information about themselves,
    remember it using the addMemory tool. When they ask questions, search your memories to provide
    personalized responses. Always be proactive about remembering important details.`
  })

  return result.toAIStreamResponse()
}

typescript

'use client'

import { useChat } from 'ai/react'

export default function PersonalAssistant() {
  const { messages, input, handleInputChange, handleSubmit } = useChat()

  return (
    <div className="flex flex-col h-screen max-w-2xl mx-auto p-4">
      <div className="flex-1 overflow-y-auto space-y-4">
        {messages.map((message) => (
          <div
            key={message.id}
            className={`p-4 rounded-lg ${
              message.role === 'user' ? 'bg-blue-100 ml-auto' : 'bg-gray-100'
            }`}
          >
            <p>{message.content}</p>
          </div>
        ))}
      </div>

      <form onSubmit={handleSubmit} className="mt-4">
        <input
          value={input}
          onChange={handleInputChange}
          placeholder="Tell me about yourself or ask me anything..."
          className="w-full p-2 border rounded"
        />
      </form>
    </div>
  )
}

</CodeGroup>

Example conversation:

User: "I'm allergic to peanuts and I love Italian food"
AI: Uses addMemory tool "I've remembered that you're allergic to peanuts and love Italian food!"
User: "Suggest a restaurant for dinner"
AI: Uses searchMemories tool "Based on what I know about you, I'd recommend an Italian restaurant that's peanut-free..."

Customer Support with Context

Build a customer support system that remembers customer history:

typescript

import { streamText } from 'ai'
import { createOpenAI } from '@ai-sdk/openai'
import { supermemoryTools } from '@supermemory/tools/ai-sdk'

const openai = createOpenAI({
  apiKey: process.env.OPENAI_API_KEY!
})

export async function POST(request: Request) {
  const { messages, customerId } = await request.json()

  const result = await streamText({
    model: openai('gpt-5'),
    messages,
    tools: supermemoryTools(process.env.SUPERMEMORY_API_KEY!, {
      containerTags: [customerId]
    }),
    system: `You are a customer support agent. Before responding to any query:
    1. Search for the customer's previous interactions and issues
    2. Remember any new information shared in this conversation
    3. Provide personalized help based on their history
    4. Always be empathetic and solution-focused`
  })

  return result.toAIStreamResponse()
}

Multi-User Learning Assistant

Build an assistant that learns from multiple users but keeps data separate:

typescript

import { streamText } from 'ai'
import { createAnthropic } from '@ai-sdk/anthropic'
import { supermemoryTools } from '@supermemory/tools/ai-sdk'

const anthropic = createAnthropic({
  apiKey: process.env.ANTHROPIC_API_KEY!
})

export async function POST(request: Request) {
  const { messages, userId, courseId } = await request.json()

  const result = await streamText({
    model: anthropic('claude-3-haiku-20240307'),
    messages,
    tools: supermemoryTools(process.env.SUPERMEMORY_API_KEY!, {
      containerTags: [userId]
    }),
    system: `You are a learning assistant. Help students with their coursework by:
    1. Remembering their learning progress and struggles
    2. Searching for relevant information from their past sessions
    3. Providing personalized explanations based on their learning style
    4. Tracking topics they've mastered vs topics they need more help with`
  })

  return result.toAIStreamResponse()
}

Research Assistant with File Processing

Combine file upload with memory tools for research assistance:

typescript

import { streamText } from 'ai'
import { createOpenAI } from '@ai-sdk/openai'
import { supermemoryTools } from '@supermemory/tools/ai-sdk'

const openai = createOpenAI({
  apiKey: process.env.OPENAI_API_KEY!
})

export async function POST(request: Request) {
  const { messages, projectId } = await request.json()

  const result = await streamText({
    model: openai('gpt-5'),
    messages,
    tools: supermemoryTools(process.env.SUPERMEMORY_API_KEY!, {
      containerTags: [projectId]
    }),
    system: `You are a research assistant. You can:
    1. Search through uploaded research papers and documents
    2. Remember key findings and insights from conversations
    3. Help synthesize information across multiple sources
    4. Track research progress and important discoveries`
  })

  return result.toAIStreamResponse()
}

typescript

import { addMemory } from '@supermemory/tools'

export async function POST(request: Request) {
  const formData = await request.formData()
  const file = formData.get('file') as File
  const projectId = formData.get('projectId') as string

  // Upload file and add to memory
  const memory = await addMemory({
    apiKey: process.env.SUPERMEMORY_API_KEY!,
    content: file, // Supermemory handles file processing
    title: file.name,
    headers: {
      'x-sm-conversation-id': projectId
    }
  })

  return Response.json({
    success: true,
    message: "Document uploaded and processed for research",
    memoryId: memory.id
  })
}

</CodeGroup>

Code Assistant with Project Memory

Create a coding assistant that remembers your codebase and preferences:

typescript

import { streamText } from 'ai'
import { createAnthropic } from '@ai-sdk/anthropic'
import {
  supermemoryTools,
  searchMemoriesTool,
  addMemoryTool
} from '@supermemory/tools/ai-sdk'

const anthropic = createAnthropic({
  apiKey: process.env.ANTHROPIC_API_KEY!
})

export async function POST(request: Request) {
  const { messages, repositoryId } = await request.json()

  const result = await streamText({
    model: anthropic('claude-3-sonnet-20240229'),
    messages,
    tools: {
      // Use individual tools for more control
      searchMemories: searchMemoriesTool(process.env.SUPERMEMORY_API_KEY!, {
        headers: {
          'x-sm-conversation-id': `repo-${repositoryId}`
        }
      }),
      addMemory: addMemoryTool(process.env.SUPERMEMORY_API_KEY!, {
        headers: {
          'x-sm-conversation-id': `repo-${repositoryId}`
        }
      }),
      // Add custom tools
      executeCode: {
        description: 'Execute code in a sandbox environment',
        parameters: z.object({
          code: z.string(),
          language: z.string()
        }),
        execute: async ({ code, language }) => {
          // Your code execution logic
          return { result: "Code executed successfully" }
        }
      }
    },
    system: `You are a coding assistant with memory. You can:
    1. Remember coding patterns and preferences from past conversations
    2. Search through previous code examples and solutions
    3. Track project architecture and design decisions
    4. Learn from debugging sessions and common issues`
  })

  return result.toAIStreamResponse()
}

Advanced: Custom Tool Integration

Combine Supermemory tools with your own custom tools:

typescript

import { streamText } from 'ai'
import { createOpenAI } from '@ai-sdk/openai'
import { supermemoryTools } from '@supermemory/tools/ai-sdk'
import { z } from 'zod'

const openai = createOpenAI({
  apiKey: process.env.OPENAI_API_KEY!
})

// Custom tool for calendar integration
const calendarTool = {
  description: 'Create calendar events',
  parameters: z.object({
    title: z.string(),
    date: z.string(),
    duration: z.number()
  }),
  execute: async ({ title, date, duration }) => {
    // Your calendar API integration
    return { eventId: "cal_123", message: "Event created" }
  }
}

export async function POST(request: Request) {
  const { messages } = await request.json()

  const result = await streamText({
    model: openai('gpt-5'),
    messages,
    tools: {
      // Spread Supermemory tools
      ...supermemoryTools(process.env.SUPERMEMORY_API_KEY!),
      // Add custom tools
      createEvent: calendarTool,
    },
    system: `You are a personal assistant that can remember information and
    manage calendars. When users mention events or appointments:
    1. Remember the details using addMemory
    2. Create calendar events using createEvent
    3. Search for conflicts using searchMemories`
  })

  return result.toAIStreamResponse()
}

Environment Setup

For all examples, ensure you have these environment variables:

bash

SUPERMEMORY_API_KEY=your_supermemory_key
OPENAI_API_KEY=your_openai_key
ANTHROPIC_API_KEY=your_anthropic_key

Best Practices

Memory Tools

Use descriptive memory content for better search results
Include context in your system prompts about when to use each tool
Use project headers to separate different use cases
Implement error handling for tool failures

General Tips

Start with simple examples and gradually add complexity
Use the search functionality to avoid duplicate memories
Implement proper authentication for production use
Consider rate limiting for high-volume applications

Next Steps

<CardGroup cols={2}> <Card title="Memory API" icon="database" href="/memory-api/overview"> Advanced memory management with full API control </Card> <Card title="User Profiles" icon="user" href="/user-profiles"> Automatic personalization with user profiles </Card> </CardGroup>