README.md

What is it?

Consoles gives AI applications access to infrastructure and enhanced capabilities through clean, intuitive APIs.

Available APIs

Extract

Transform any content into structured data with AI.

Installation

npm install consoles-ai

Quick Start

import { Console } from 'consoles-ai';
import { z } from 'zod';

const consoles = new Console('your-api-key');

// Using Zod schema with descriptions
const productSchema = z.object({
  name: z.string().describe('Product name'),
  price: z.number().describe('Price in USD'),
  description: z.string().describe('Product description'),
  inStock: z.boolean().optional().describe('Current availability status')
});

const data = await consoles.extract({
  type: 'text',
  content: `
    New iPhone 15 Pro
    Price: $999
    Experience the most powerful iPhone ever.
    Currently in stock.
  `,
  schema: productSchema
});

// Response:
// {
//   status: 'success',
//   result: {
//     name: "iPhone 15 Pro",
//     price: 999,
//     description: "Experience the most powerful iPhone ever",
//     inStock: true
//   },
//   usage: {
//     input_tokens: 28,
//     output_tokens: 12,
//     total_tokens: 40,
//     input_cost: "0.0003",
//     output_cost: "0.0001",
//     total_cost: "0.0004"
//   }
// }

Examples

Document Analysis

const financials = await consoles.extract({
  type: 'url',
  content: 'https://s22.q4cdn.com/959853165/files/doc_financials/2023/ar/NVDA-2023-Annual-Report.pdf',
  schemaDescription: "Extract financial metrics including revenue, net income, and GPU revenue (all in millions USD)'"
  prompt: 'Extract the key financial metrics from FY2023'
});

// Response
{
  status: 'success',
  result: {
    revenue: 26974,
    netIncome: 4368,
    gpuRevenue: 22035
  },
  usage: {
    input_tokens: 4537,
    output_tokens: 21,
    total_tokens: 4558,
    input_cost: "0.0045",
    output_cost: "0.0013",
    total_cost: "0.0058"
  }
}

Media Processing

// Extract podcast insights with streaming
const podcastSchema = z.object({
  topics: z.array(z.string()).describe('Main topics discussed in the podcast'),
  keyMoments: z.array(
    z.object({
      timestamp: z.string().describe('Timestamp in HH:MM:SS format'),
      summary: z.string().describe('Brief summary of the key moment')
    })
  )
});

const stream = await consoles.extract({
  type: 'file',
  content: new Blob([audioBuffer], { type: 'audio/mp3' }),
  schema: podcastSchema,
  stream: true
});

// Process stream chunks
for await (const chunk of stream) {
  console.log('Received chunk:', chunk);
}

// Generate YouTube chapters
const chapterSchema = z.array(
  z.object({
    timestamp: z.string().describe('Chapter start time in HH:MM:SS format'),
    title: z.string().describe('Short, descriptive chapter title'),
    summary: z.string().describe('Detailed description of chapter content')
  })
);

const chapters = await consoles.extract({
  type: 'url',
  content: 'https://youtube.com/watch?v=example',
  schema: chapterSchema,
  prompt: 'Based on the video provided, Generate detailed chapter markers with timestamps and summaries'
});

// Response
{
  status: 'success',
  result: [
    {
      timestamp: "00:00:00",
      title: "Introduction",
      summary: "Overview of topics and speaker introduction"
    },
    {
      timestamp: "00:02:15",
      title: "Core Concepts",
      summary: "Explanation of fundamental principles and key terminology"
    },
    {
      timestamp: "00:15:30",
      title: "Real World Applications",
      summary: "Practical examples and use cases in production environments"
    }
  ],
  usage: {
    input_tokens: 3245,
    output_tokens: 156,
    total_tokens: 3401,
    input_cost: "0.0032",
    output_cost: "0.0011",
    total_cost: "0.0043"
  }
}

Using Natural Language Schema

// Extract with schema description
const data = await consoles.extract({
  type: 'file',
  content: {
    data: pdfBuffer.toString('base64'),
    mimeType: 'application/pdf'
  },
  schemaDescription: 'Extract company metrics including revenue, profit margins, and year-over-year growth. Revenue should be a number, margins should be percentages, and growth should be a number representing the percentage change.',
  prompt: 'Focus on the most recent fiscal year'
});

// Response
{
  status: 'success',
  result: {
    revenue: 26974000000,
    profitMargin: 28.5,
    yearOverYearGrowth: 32.7
  },
  usage: {
    input_tokens: 3245,
    output_tokens: 156,
    total_tokens: 3401,
    input_cost: "0.0032",
    output_cost: "0.0011",
    total_cost: "0.0043"
  }
}

Browser

Managed browser automation at scale.

Computer

On-demand compute resources for AI workloads.

Documentation

Visit our documentation for more information:

• API Reference - Complete API documentation including supported file types and formats
• Tutorials - Step-by-step guides to get started
• How-to Guides - Practical guides for common tasks
• Explanation - Concepts and technical details

Links

Docs • Discord • [email protected]