LlamaIndex.TS Setup

Install and configure LlamaIndex.TS, a lightweight data framework for integrating large language models with your own data in JavaScript/TypeScript environments.

What This Skill Does

This skill helps you set up LlamaIndex.TS in your project, including:

Instructions

Follow these steps to set up LlamaIndex.TS:

1. Verify Runtime Compatibility

Check that the project uses a supported JavaScript runtime:

Note: Browser support is currently limited due to AsyncLocalStorage API requirements.

2. Install Core Package

Install the llamaindex package using the project's package manager:

```bash

npm install llamaindex

or

pnpm install llamaindex

or

yarn add llamaindex

```

3. Identify Required Provider Packages

Ask the user which providers they need, or analyze existing code to determine requirements:

**LLM Providers** (choose one or more):

**Other Providers**:

4. Install Provider Packages

Install the identified provider packages. For example, for OpenAI:

```bash

npm install @llamaindex/openai

```

5. Verify TypeScript Configuration

If the project uses TypeScript, ensure `tsconfig.json` is properly configured:

6. Create Basic Setup Example

Provide a minimal working example based on the installed providers. For OpenAI:

```typescript

import { OpenAI } from "@llamaindex/openai";

import { Document, VectorStoreIndex } from "llamaindex";

// Initialize LLM

const llm = new OpenAI({ model: "gpt-4", apiKey: process.env.OPENAI_API_KEY });

// Create documents

const document = new Document({ text: "Your text here" });

// Create index and query

const index = await VectorStoreIndex.fromDocuments([document]);

const queryEngine = index.asQueryEngine({ llm });

const response = await queryEngine.query("Your question here");

```

7. Add Environment Variables

Remind the user to add necessary API keys to their environment:

```bash

OPENAI_API_KEY=your_api_key_here

Add other provider keys as needed

```

8. Reference Documentation

Point the user to key resources:

Example Usage

**User Request**: "Set up LlamaIndex with OpenAI for my Next.js app"

**Agent Actions**:

1. Verify Node.js version is >= 20

2. Install `llamaindex` and `@llamaindex/openai`

3. Check for existing `tsconfig.json` configuration

4. Create example code showing basic document indexing and querying

5. Add `OPENAI_API_KEY` to `.env.local` (if not present)

6. Provide link to Next.js-specific documentation

Important Notes

Constraints