AI Data Extraction
Turn any webpage into structured JSON. AI-powered extraction that understands meaning, not just HTML. Define your schema, get structured data.
$2.00 / 1,000 extractions
- Auto Mode
- Schema Mode
- Markdown Mode
// Extract product data with schemaconst product = await stack0.extraction.extractAndWait({url: 'https://store.example.com/product/123',mode: 'schema',schema: {type: 'object',properties: {name: { type: 'string' },price: { type: 'number' },inStock: { type: 'boolean' },rating: { type: 'number' },},},})console.log(product.extractedData)// { name: '...', price: 29.99, inStock: true, rating: 4.5 }
Built for production
Semantic Understanding
AI understands content meaning, not just HTML structure.
{mode: 'auto' // AI determines what's important,}
AI understands content meaning, not just HTML structure.
Schema Validation
JSON Schema support ensures you always get the structure you expect.
{schema: { type: 'object', properties: { ... } },}
JSON Schema support ensures you always get the structure you expect.
Batch Processing
Extract from multiple URLs with shared schema in parallel.
await stack0.extraction.batchAndWait({urls,config,})
{id: "job_1gcdw",status: "200 OK",createdAt: "2026-09-02T14:03:06Z",}
Dynamic Content
Handle SPAs and lazy-loaded content with wait conditions.
{waitForSelector: '.content-loaded',}
Handle SPAs and lazy-loaded content with wait conditions.
Guided Extraction
Add prompts to focus AI on specific content or filter results.
{prompt: 'Extract only technical team members',}
Add prompts to focus AI on specific content or filter results.
Webhook Support
Start async extractions and receive results via webhook.
{webhookUrl: 'https://yourapp.com/webhook',}
Start async extractions and receive results via webhook.
Auto Mode
AI automatically identifies and extracts the most relevant content.
Schema Mode
Define your structure with JSON Schema. Get consistent, typed output.
Markdown Mode
Convert page content to clean, formatted markdown with links.
HTML Mode
Get raw HTML for custom parsing when you need full control.
Custom Prompts
Guide extraction with natural language instructions.
No Maintenance
No CSS selectors to update when sites change their HTML.
Built for real applications
RAG Data Pipelines
Feed structured web content into retrieval-augmented generation systems.
Simple, usage-based
$20/mo includes $10 of usage. Metered beyond that, at the same rate for everyone.
Extractions
$2.00/ 1,000 extractions
- All extraction modes included
- AI tokens included in price
- JSON Schema validation
- Custom prompts
- Page metadata included
- Batch processing
- Webhook notifications
Frequently asked questions
Traditional scraping uses CSS selectors that break when sites change. AI extraction understands content semantically—it finds the price whether it's in a <span>, <div>, or <p>. Your schema defines what you want, not where to find it.
Four modes: Auto (AI determines what's important), Schema (you define the structure), Markdown (clean formatted text), and HTML (raw content for custom parsing).
Yes. Add a prompt to any extraction to focus the AI on specific content, filter results, or provide context about what you're looking for.
Use waitForSelector to wait for elements to appear, or waitForTimeout for a delay. This handles SPAs, lazy-loading, and JavaScript-rendered content.
Yes. The $2/1K extractions price includes all AI processing. No separate token charges.
Yes. Use batch extraction to process multiple URLs with a shared schema. Results are processed in parallel with webhook notifications.
Ready to build?
Get started in minutes.