LlamaIndex

Show me reference photos for winter trips in Japan with the help of LlamaIndex and Vity Toolkit in less than 15 lines

In this guide you'll learn how to implement and use Vity Toolkit with LlamaIndex in just a few lines of code. For the purpose of this example, we'll create an assistant that can searches for photos using Pexels.

Show me reference photos for winter trips in Japan

In this example, we will use LlamaIndex to searches for photos from Pexels using Vity Toolkit.

1

Install the package.

npm install vity-toolkit @llamaindex/openai @llamaindex/workflow

2

Initialize Vity Toolkit.

import { agent } from "@llamaindex/workflow";
import { openai } from "@llamaindex/openai";
import { App, LlamaIndexToolkit } from "vity-toolkit";

const P = (globalThis as any).process;
const apiKey = P?.env?.OPENAI_API_KEY;

const toolkit = new LlamaIndexToolkit({
    userPrivateKey: P?.env?.SOLANA_PRIVATE_KEY,
    appPrivateKey: P?.env?.SOLANA_PRIVATE_KEY,
});

3

Get the required tools.

const tools = toolkit.getTools({ apps: [App.PEXELS] });

4

Define the Agent.

const pexelsAgent = agent({
    name: "Pexels Agent",
    description: "An agent that searches for photos using Pexels",
    llm: openai({ model: "gpt-4o-mini", apiKey }),
    systemPrompt: "You are a travel assistant that curates inspiring visuals. Use the available tools to search for photos.",
    tools,
});

5

Execute the Agent.

const result = await pexelsAgent.run("Show me reference photos for winter trips in Japan.");

console.log("Agent result:", result.data.message.content);

vity-toolkit/code-samples/agent/llamaindex.ts at main · apneduniya/vity-toolkitGitHub

Full code