AI Extraction

import { Stack0 } from '@stack0/sdk'
 
const stack0 = new Stack0({
  apiKey: process.env.STACK0_API_KEY!
})
 
// Start extraction (async)
const { id, status } = await stack0.extraction.extract({
  url: 'https://example.com/article',
  mode: 'markdown',
  includeMetadata: true,
})
 
// Poll for completion
const extraction = await stack0.extraction.get({ id })
console.log('Markdown:', extraction.markdown)
console.log('Metadata:', extraction.pageMetadata)

// This method handles polling automatically
const extraction = await stack0.extraction.extractAndWait({
  url: 'https://example.com/article',
  mode: 'markdown',
  includeMetadata: true,
  includeLinks: true,
  includeImages: true,
}, {
  pollInterval: 1000, // Check every 1 second (default)
  timeout: 60000,     // Timeout after 60 seconds (default)
})
 
console.log('Title:', extraction.pageMetadata?.title)
console.log('Description:', extraction.pageMetadata?.description)
console.log('Links found:', extraction.pageMetadata?.links?.length)
console.log('Markdown content:', extraction.markdown)

// Define a schema for the data you want to extract
const extraction = await stack0.extraction.extractAndWait({
  url: 'https://example.com/product/iphone-15',
  mode: 'schema',
  schema: {
    type: 'object',
    properties: {
      name: { type: 'string', description: 'Product name' },
      price: { type: 'number', description: 'Price in USD' },
      currency: { type: 'string' },
      description: { type: 'string' },
      rating: { type: 'number', description: 'Average rating out of 5' },
      reviewCount: { type: 'number' },
      availability: {
        type: 'string',
        enum: ['in_stock', 'out_of_stock', 'preorder']
      },
      features: {
        type: 'array',
        items: { type: 'string' }
      },
      images: {
        type: 'array',
        items: { type: 'string', description: 'Image URL' }
      }
    },
    required: ['name', 'price']
  }
})
 
// extractedData matches your schema
const product = extraction.extractedData as {
  name: string
  price: number
  currency: string
  description: string
  rating: number
  reviewCount: number
  availability: string
  features: string[]
  images: string[]
}
 
console.log(`${product.name}: $${product.price}`)
console.log(`Rating: ${product.rating}/5 (${product.reviewCount} reviews)`)

const extraction = await stack0.extraction.extractAndWait({
  url: 'https://news.example.com/article/tech-layoffs',
  mode: 'schema',
  // Natural language prompt to guide extraction
  prompt: `Extract the main article content. Focus on:
    - The main headline and any subheadings
    - Key facts and figures mentioned
    - Quotes from sources with attribution
    - The publication date and author
    Ignore ads, navigation, and related articles.`,
  schema: {
    type: 'object',
    properties: {
      headline: { type: 'string' },
      author: { type: 'string' },
      publishDate: { type: 'string' },
      summary: { type: 'string', description: '2-3 sentence summary' },
      keyFacts: {
        type: 'array',
        items: { type: 'string' }
      },
      quotes: {
        type: 'array',
        items: {
          type: 'object',
          properties: {
            text: { type: 'string' },
            source: { type: 'string' }
          }
        }
      }
    }
  }
})

const extraction = await stack0.extraction.extractAndWait({
  url: 'https://app.example.com/dashboard',
  mode: 'markdown',
  // Set authentication headers
  headers: {
    'Authorization': 'Bearer your-token',
  },
  // Or use cookies
  cookies: [
    { name: 'session', value: 'abc123', domain: 'app.example.com' },
  ],
  // Wait for dynamic content to load
  waitForSelector: '.dashboard-content',
})

// Start a batch extraction job
const batch = await stack0.extraction.batch({
  urls: [
    'https://example.com/article/1',
    'https://example.com/article/2',
    'https://example.com/article/3',
  ],
  config: {
    mode: 'markdown',
    includeMetadata: true,
  },
})
 
console.log('Batch ID:', batch.id)
console.log('Total URLs:', batch.totalUrls)
 
// Wait for completion
const completedJob = await stack0.extraction.batchAndWait({
  urls: [
    'https://example.com/product/1',
    'https://example.com/product/2',
  ],
  config: {
    mode: 'schema',
    schema: {
      type: 'object',
      properties: {
        name: { type: 'string' },
        price: { type: 'number' }
      }
    }
  }
}, {
  pollInterval: 2000,
  timeout: 300000, // 5 minutes
})
 
console.log(`Completed: ${completedJob.successfulUrls} success, ${completedJob.failedUrls} failed`)

// Create a daily extraction schedule
const schedule = await stack0.extraction.createSchedule({
  name: 'Daily price monitoring',
  url: 'https://competitor.com/pricing',
  frequency: 'daily',
  config: {
    mode: 'schema',
    schema: {
      type: 'object',
      properties: {
        plans: {
          type: 'array',
          items: {
            type: 'object',
            properties: {
              name: { type: 'string' },
              price: { type: 'number' },
              features: { type: 'array', items: { type: 'string' } }
            }
          }
        }
      }
    }
  },
  // Get notified when content changes
  detectChanges: true,
  webhookUrl: 'https://your-app.com/webhook',
})
 
// List extraction schedules
const { items } = await stack0.extraction.listSchedules({
  isActive: true,
})

// List extractions with filters
const { items, nextCursor } = await stack0.extraction.list({
  status: 'completed',
  limit: 20,
})
 
for (const extraction of items) {
  console.log(`[${extraction.status}] ${extraction.url}`)
  console.log(`  Mode: ${extraction.mode}`)
  console.log(`  Tokens used: ${extraction.tokensUsed}`)
}
 
// Get a specific extraction
const extraction = await stack0.extraction.get({ id: 'extraction-id' })
 
// Delete an extraction
await stack0.extraction.delete({ id: 'extraction-id' })

const usage = await stack0.extraction.getUsage({
  periodStart: '2024-01-01T00:00:00Z',
  periodEnd: '2024-01-31T23:59:59Z',
})
 
console.log('Extractions:', usage.extractionsTotal)
console.log('  Successful:', usage.extractionsSuccessful)
console.log('  Failed:', usage.extractionsFailed)
console.log('  Tokens used:', usage.extractionTokensUsed)
console.log('  Credits used:', usage.extractionCreditsUsed)