To use Scoopika in web applications, both server-side and client-side, we recommend creating a Scoopika endpoint. This setup can be done in under 5 minutes.

Why Create a Scoopika Endpoint?

Creating a Scoopika endpoint simplifies the integration process and enhances the functionality of your AI agents. Here’s why:

  1. Ease of Use: Instead of managing HTTP requests and streaming responses manually, the Scoopika endpoint handles all communication between the client and server seamlessly. This makes client-side usage as straightforward as server-side usage.
  2. Enhanced Functionality: The endpoint allows you to add tools to your agents from both the server and the client-side, unlocking a wide range of custom functionalities.
  3. Flexibility: It works out of the box with any web framework, providing a versatile solution for various application needs.


A Scoopika API endpoint offers several built-in features:

  • Client-Side Usage: Enables all Scoopika functionalities on the client-side, including agent interactions, data loading, and session management.
  • Server-Client Streaming: Supports real-time streaming hooks on the client-side for agent response tokens, tool calls, audio, and other events.
  • Caching: Caches agent data on your servers to improve performance and reduce load usage.
  • Client-Side Actions: Allows agents to execute tools safely in the user’s browser in real-time.

Example usage (with express)

Here’s a simple example of setting up a Scoopika endpoint using Express


npm install @scoopika/scoopika
import express from "express";
import { Scoopika, Endpoint } from "@scoopika/scoopika";

// Setup Scoopika
const scoopika = new Scoopika(); // requires SCOOPIKA_TOKEN env variable
const endpoint = new Endpoint({
    agents: ["AGENT_ID_1"] // List of agent IDs that can be used through this endpoint

const app = express();
app.use(express.json()); // required middleware

// Define the Scoopika endpoint"/scoopika", (req, res) => {
        request: req.body,
        stream: (s) => res.write(s),
        end: () => res.end()

app.listen(3000, () => {
    console.log('Server is running on port 3000');


npm install @scoopika/client
import { Client, Agent } from "@scoopika/client";

const client = new Client("API_URL/scoopika");
const agent = new Agent(client, "AGENT_ID");

(async () => {
    const res = await{
        inputs: { message: "Hello" },
        hooks: {
            onToken: (t) => console.log(t) // Streaming hooks


Learn more

For detailed instructions on using Scoopika with web applications and adding tools to agents on the server-side, see the full guide here.