deca-fix

DecaFix

A utility that provides AI-powered error fixing suggestions for your JavaScript/TypeScript code.

Installation

npm install deca-fix

Setup

You'll need to set up environment variables for the API key. Create a .env file in your project root:

GROQ_API_KEY=your_groq_api_key_here

or whatever provider you are using

Usage

CommonJS (Node.js scripts)

require('dotenv').config();
const { DecaFix } = require('deca-fix');

// Initialize with your API configuration
const decaFix = new DecaFix({
  apiKey: process.env.GROQ_API_KEY,
  baseUrl: "https://api.groq.com/openai/v1", // or use your own api url for private api
  model: "meta-llama/llama-4-maverick-17b-128e-instruct"
});

// Use in a try/catch block to get AI-powered fix suggestions
async function main() {
  try {
    // Your code that might throw an error
    const fs = require('fs');
    const content = fs.readFileSync('non-existent-file.txt', 'utf8');
    console.log(content);
  } catch (error) {
    // DecaFix will analyze the error and suggest a solution
    await decaFix.suggestFix(error);
  }
}

main();

ES Modules (React, Next.js, etc.)

import { config } from 'dotenv';
import { DecaFix } from 'deca-fix';

// Load environment variables
config();

// Initialize with your API configuration
const decaFix = new DecaFix({
  apiKey: process.env.GROQ_API_KEY,
  baseUrl: "https://api.groq.com/openai/v1",
  model: "meta-llama/llama-4-maverick-17b-128e-instruct"
});

// Example usage in a React component
function MyComponent() {
  const handleApiCall = async () => {
    try {
      // Your code that might throw an error
      const response = await fetch('/api/non-existent-endpoint');
      const data = await response.json();
    } catch (error) {
      // Get AI-powered fix suggestions
      await decaFix.suggestFix(error);
    }
  };

  return (
    <button onClick={handleApiCall}>Make API Call</button>
  );
}

export default MyComponent;

How It Works

DecaFix uses LLMs (specifically Llama 4 Maverick by default) to analyze JavaScript/TypeScript errors and provide code fixes. It:

• Captures the error and its stack trace
• Identifies the most relevant part of the stack trace
• Uses AI to generate a suggested fix
• Outputs the suggestion to the console

Configuration Options

When initializing DecaFix, you can provide the following options:

const decaFix = new DecaFix({
  apiKey: process.env.GROQ_API_KEY,  // Required
  baseUrl: "https://api.groq.com/openai/v1",  // API endpoint
  model: "meta-llama/llama-4-maverick-17b-128e-instruct",  // LLM model to use
  // Any other options supported by deca-chat
});

Dependencies

• deca-chat: The base chat interface used by DecaFix
• dotenv: For loading environment variables

License

MIT