Integration Tutorial April 13, 2026

REM Labs Node.js SDK: Complete Guide

The @remlabs/sdk package gives any Node.js or TypeScript application persistent AI memory. Fully typed, promise-based, and compatible with ESM and CommonJS. This guide covers everything from basic usage to advanced patterns like streaming, batching, and framework integrations.

Installation

npm install @remlabs/sdk
# or
yarn add @remlabs/sdk
# or
pnpm add @remlabs/sdk

Quick Start

import { RemLabs } from "@remlabs/sdk";

const client = new RemLabs({ apiKey: "your-api-key" });

// Store a memory
await client.remember({
  content: "The app uses Next.js 14 with App Router",
  tags: ["stack", "frontend"],
  namespace: "my-project"
});

// Search memories
const results = await client.recall({
  query: "what frontend framework do we use",
  namespace: "my-project",
  limit: 5
});

console.log(results[0].content);
// → "The app uses Next.js 14 with App Router"

// Delete a memory
await client.forget({ memoryId: results[0].id });

TypeScript Types

The SDK exports full TypeScript types for all parameters and return values:

import {
  RemLabs,
  RememberParams,
  RecallParams,
  Memory,
  RecallResult
} from "@remlabs/sdk";

const params: RememberParams = {
  content: "API uses REST with JSON responses",
  tags: ["api", "conventions"],
  namespace: "backend"
};

await client.remember(params);

const results: RecallResult = await client.recall({
  query: "API design",
  limit: 10
});

results.forEach((mem: Memory) => {
  console.log(mem.id, mem.content, mem.score);
});

Express.js Integration

import express from "express";
import { RemLabs } from "@remlabs/sdk";

const app = express();
const memory = new RemLabs({
  apiKey: process.env.REMLABS_API_KEY
});

app.post("/chat", express.json(), async (req, res) => {
  const { userId, message } = req.body;

  // Recall user context
  const context = await memory.recall({
    query: message,
    namespace: `user-${userId}`,
    limit: 5
  });

  // ... generate response with your LLM ...

  // Store the exchange
  await memory.remember({
    content: `User: ${message}\nBot: ${reply}`,
    namespace: `user-${userId}`,
    tags: ["conversation"]
  });

  res.json({ reply });
});

Next.js API Route

// app/api/memory/route.ts
import { RemLabs } from "@remlabs/sdk";
import { NextRequest, NextResponse } from "next/server";

const memory = new RemLabs({
  apiKey: process.env.REMLABS_API_KEY!
});

export async function POST(req: NextRequest) {
  const { action, ...params } = await req.json();

  if (action === "remember") {
    await memory.remember(params);
    return NextResponse.json({ ok: true });
  }

  if (action === "recall") {
    const results = await memory.recall(params);
    return NextResponse.json({ memories: results });
  }

  return NextResponse.json({ error: "Unknown action" }, {
    status: 400
  });
}

Batch Operations

Store multiple memories efficiently with batch operations:

// Batch remember
await client.rememberBatch([
  { content: "DB uses PostgreSQL", tags: ["db"] },
  { content: "Cache layer is Redis", tags: ["cache"] },
  { content: "Queue is RabbitMQ", tags: ["queue"] }
], { namespace: "infrastructure" });

Environment Variables

The SDK reads REMLABS_API_KEY from process.env if no key is passed:

// .env
REMLABS_API_KEY=your-api-key

// Then:
const client = new RemLabs(); // reads from env

Edge compatible: The SDK uses the Fetch API internally, making it compatible with Vercel Edge Functions, Cloudflare Workers, and Deno Deploy.

Add memory to your Node.js app

npm install @remlabs/sdk. Free tier. Full TypeScript support.

Get started free →