nextstring

nextstring is a powerful TypeScript library that extends the native String prototype with AI-driven capabilities. It allows developers to perform advanced operations on strings, such as extracting data, answering questions, summarizing text, checking conditions, and translating text, all powered by a configurable AI provider.

Features

• Extract Data: Extract specific information from a string based on descriptions.
• Answer Questions: Answer questions using the context of the string.
• Summarize: Summarize text to a specified number of words.
• Check Conditions: Evaluate conditions based on the string's content.
• Translate: Translate text into a specified language.

Installation

npm install nextstring

Usage

Initialization

Before using the library, initialize it with an AI provider:

import { initialise, OpenaiProvider } from "nextstring";

const provider = new OpenaiProvider({apiKey: "your-api-key"});
initialise(provider);

Examples

1. Extracting Data

const data = "John's email is [e-mail] and his phone number is 123-456-7890.";
const items = [
    { name: "email", description: "The email address in the data" },
    { name: "phone", description: "The phone number in the data" },
];

const result = await data.extractData(items);
console.log(result); 
// Output: { email: "[e-mail]", phone: "123-456-7890" }

2. Answering Questions

const text = "The Eiffel Tower is located in Paris, France.";
const question = "Where is the Eiffel Tower located?";

const answer = await text.question(question);
console.log(answer); 
// Output: "Paris, France"

3. Summarizing Text

const longText = "This is a long paragraph that needs to be summarized into fewer words for better readability.";
const summary = await longText.summarise(10);
console.log(summary); 
// Output: "This is a long paragraph summarized for readability."

4. Checking Conditions

const content = "The weather today is sunny and warm.";
const condition = "Is the weather sunny?";

const isSunny = await content.checkIf(condition);
console.log(isSunny); 
// Output: true

5. Translating Text

const text = "Hello, how are you?";
const translatedText = await text.translate("spanish");
console.log(translatedText); 
// Output: "Hola, ¿cómo estás?"

6. Rewriting Text

const data = "This is some original text.";
const instructions = "Rewrite this to be more formal.";

const rewritten = await data.rewrite(instructions);
console.log(rewritten); 
// Output: "This is some original text rewritten formally."

7. Classifying Text

const text = "This is a message about a technical issue.";
const categories = [
    { name: "support", description: "Messages related to customer support" },
    { name: "sales", description: "Messages related to sales inquiries" }
];

const result = await text.classifyText(categories);
console.log(result); 
// Output: "support"

Impact on AI Coding

nextstring simplifies the integration of AI capabilities into your applications by abstracting complex AI interactions into intuitive string methods. This allows developers to:

• Focus on building features rather than implementing AI logic.
• Leverage AI for natural language processing tasks with minimal effort.
• Enhance productivity and reduce boilerplate code.

Sample Scenario: AI-Powered Customer Support

Imagine you're building a customer support chatbot. With nextstring, you can easily extract key information from user messages, answer their questions, summarize lengthy responses, and translate messages for multilingual support.

const userMessage = "Hi, my name is Alice. My order ID is 12345, and I need help with a refund.";
const items = [
    { name: "name", description: "The user's name" },
    { name: "orderID", description: "The order ID mentioned in the message" },
];

const extractedData = await userMessage.extractData(items);
console.log(extractedData); 
// Output: { name: "Alice", orderID: "12345" }

const question = "What does the user need help with?";
const userNeed = await userMessage.question(question);
console.log(userNeed); 
// Output: "a refund"

const translatedMessage = await userMessage.translate("spanish");
console.log(translatedMessage);
// Output: "Hola, mi nombre es Alice. Mi ID de pedido es 12345 y necesito ayuda con un reembolso."

With just a few lines of code, you can build a robust AI-powered system to handle customer queries efficiently.

Contributing

Contributions are welcome! Please open an issue or submit a pull request for any improvements or bug fixes.

License

This project is licensed under the MIT License.