meloqui

Examples

This directory contains example code demonstrating how to use the Agent Chat SDK.

Prerequisites

Before running the examples, you need to:

1. Install dependencies: npm install
2. Build the project: npm run build
3. Set up your API keys (see below)

Setting Up API Keys

OpenAI

Set the OPENAI_API_KEY environment variable:

export OPENAI_API_KEY=your-openai-api-key-here

Running Examples

Basic Chat

Demonstrates simple chat interaction with OpenAI:

npm run example:basic

This example shows:

• Creating a ChatClient with OpenAI provider
• Sending a simple message
• Receiving and displaying the response
• Accessing response metadata (model, tokens, etc.)

Streaming

Real-time response streaming:

npm run example:streaming

This example demonstrates:

• Streaming responses in real-time
• Processing chunks as they arrive
• Building complete responses incrementally

Conversation History

Multi-turn conversations with automatic history:

npm run example:history

This example demonstrates:

• Using conversationId for persistent conversations
• Automatic message history tracking
• Retrieving full conversation history
• Context preservation across multiple turns

Persistent Storage

File-based storage for conversation persistence:

npm run example:storage

This example demonstrates:

• Using FileStorage for persistent conversations
• Saving conversations to disk
• Loading conversations across sessions
• Maintaining context between application restarts

Example Files

• basic-chat.ts - Simple chat interaction with OpenAI
• streaming.ts - Real-time streaming responses
• with-history.ts - Multi-turn conversations with history
• with-storage.ts - Persistent storage across sessions

Coming Soon

Additional examples will be added in future releases:

• Multiple providers (Anthropic, Google, Ollama)
• Error handling and retries
• Rate limiting
• Custom configurations

This site is open source. Improve this page.