@abstraxn/chat-ai
v0.0.1
Published
Abstraxn ChatAI package for Simple and efficient communication with ChatAI API
Readme
ChatAI
Overview
The Abstraxn ChatAI SDK is a simple and efficient SDK for communicating with the ChatAI API. This SDK provides easy-to-use methods for both streaming and non-streaming chat interactions.
Installation
npm install @abstraxn/chat-aiOr using yarn package manager:
yarn add @abstraxn/chat-aiQuick Start
import { ChatAIClient } from "@abstraxn/chat-ai";
const client = new ChatAIClient({
apiKey: "your-api-key-here",
});
// Simple chat
const response = await client.chat("What is machine learning?");
console.log(response.response);Features
- Simple API - Easy to use with minimal configuration
- Streaming Support - Real-time streaming responses with Server-Sent Events
- Session Management - Maintain conversation context across requests
- CORS Support - Configurable headers for cross-origin requests
- TypeScript Support - Full type definitions included
- Error Handling - Comprehensive error handling with detailed messages
- Custom Headers - Support for additional HTTP headers
Usage Examples
Basic Chat (Non-Streaming)
import { ChatAIClient } from "@abstraxn/chat-ai";
const client = new ChatAIClient({
apiKey: "your-api-key",
});
async function basicChat() {
try {
const response = await client.chat("What is artificial intelligence?");
console.log("AI Response:", response.response);
console.log("Session ID:", response.sessionId);
} catch (error) {
console.error("Error:", error.message);
}
}
basicChat();Streaming Chat
import { ChatAIClient } from "@abstraxn/chat-ai";
const client = new ChatAIClient({
apiKey: "your-api-key",
});
async function streamingChat() {
try {
const stream = await client.chatStream("Explain deep learning in detail");
for await (const chunk of stream) {
if (chunk.type === "content") {
process.stdout.write(chunk.content); // Print streaming content
}
if (chunk.type === "done") {
console.log("\n[Stream complete]");
}
}
} catch (error) {
console.error("Streaming error:", error.message);
}
}
streamingChat();Session Management
import { ChatAIClient } from "@abstraxn/chat-ai";
const client = new ChatAIClient({ apiKey: "your-api-key" });
async function conversation() {
const sessionId = "user-session-123";
// First message
const response1 = await client.chat("Hello, my name is John", {
sessionId: sessionId,
});
console.log("AI:", response1.response);
// Follow-up message (AI will remember the conversation)
const response2 = await client.chat("What is my name?", {
sessionId: sessionId,
});
console.log("AI:", response2.response);
}Custom Headers and CORS
const client = new ChatAIClient({
apiKey: "your-api-key",
origin: "https://my-frontend.com",
headers: {
"X-Request-ID": "abc123",
"X-Client-Version": "1.0.0",
},
});Error Handling
try {
const response = await client.chat("Tell me about AI");
console.log(response.response);
} catch (error) {
if (error.status) {
console.error(`API Error ${error.status}:`, error.message);
} else {
console.error("Network Error:", error.message);
}
}API Reference
ChatAIClient
Constructor
new ChatAIClient(config);Config Options:
apiKey(required): Your ChatAI API keytimeout(optional): Request timeout in milliseconds (default:30000)origin(optional): Origin header for CORS (default:*)headers(optional): Additional custom headers
Methods
chat(query, options)
Send a chat message and get a complete response.
Parameters:
query(string): The message to sendoptions(object, optional):sessionId(string, optional): Session ID for conversation continuitystream(boolean, optional): Force non-streaming mode (default:false)
Returns: Promise
chatStream(query, options)
Send a chat message and get a streaming response.
Parameters:
query(string): The message to sendoptions(object, optional):sessionId(string, optional): Session ID for conversation continuity
Returns: AsyncGenerator
Types
ChatResponse
interface ChatResponse {
error: boolean;
sessionId: string;
response: string;
outOfContext?: boolean;
}ChatChunk
interface ChatChunk {
type: "session" | "content" | "done" | "error";
sessionId?: string;
content?: string;
timestamp?: string;
timing?: {
total: number;
};
outOfContext?: boolean;
message?: string;
}