Skip to content

About

Pre-defined extraction schemas for high-value, high-scale data use cases powered by Tabstack

Resources

Stars

4 stars

Watchers

0 watching

Forks

Repository files navigation

Tabstack Schemas

Ready-to-use JSON Schema definitions for Tabstack's extract.json. Point one at any URL and get structured, typed data back. 46 schemas across 10 categories.

Grab a file straight from this repo (or the docs site) and pass it to the API.

Use it

import Tabstack from '@tabstack/sdk'
import schema from './real-estate/residential-listing.json' with { type: 'json' }

const client = new Tabstack() // reads TABSTACK_API_KEY from the environment

const url = 'https://www.example.com/listing/123'
const data = await client.extract.json({
  url,
  json_schema: schema,
  effort: 'standard', // 'min' | 'standard' | 'max'
})

// page_title and favicon are auto-filled by Tabstack from page metadata.
// Stamp request-side provenance yourself so it is always correct:
const record = { ...data, source_url: url, extracted_at: new Date().toISOString() }

Python and the other SDKs take the same json_schema object.

Categories

Category Schemas Includes
b2b-intel 7 API Documentation Endpoint, Company Profile, Funding Round, Press Release & News Mention, Product Changelog Entry, SaaS Pricing Page, Tech Stack
developer-ecosystem 2 GitHub Repository, Package Registry Entry
e-commerce 5 Category Search Results, Grocery & CPG Product, Marketplace Seller Profile, Product Listing, Product Review
finance 5 Alternative Asset Listing, Angel & Seed Deal, Crypto Asset, Public Company Filing, VC Portfolio Page
gov-public-records 5 Building Permit, Business Registration, Court Filing / Case, Government Contract Award, Property Tax Record
healthcare 4 Clinical Trial, Drug Formulary Entry, Medical Device Recall, Provider Directory Entry
jobs 5 Company Hiring Velocity, Executive Leadership Change, Job Posting, Layoff Event, Salary Data Point
real-estate 6 Commercial Property, Foreclosure & Auction Listing, Property Sold History, Rental Listing, Residential Listing, Short-Term Rental
social-content 3 News Article, Social Profile, Software Review
travel 4 Attraction & Experience, Flight Price Snapshot, Hotel Listing, Restaurant Listing

Each category folder has its own README with a runnable example. index.json is a machine-readable manifest of every schema.

How these schemas are built

  • Standard JSON Schema (draft-07). Every field has a description that also notes where the value usually appears on the page, which steers extraction.
  • Closed enums include an "other" value so off-list page wording is captured instead of dropped.
  • Missing fields come back as null rather than failing the call, so partial pages still return useful data.
  • Provenance (source_url, extracted_at) is left to the caller, which keeps it accurate. page_title and favicon are auto-filled by Tabstack when present.

Structure

<category>/
  README.md
  <schema-name>.json
index.json

About

Pre-defined extraction schemas for high-value, high-scale data use cases powered by Tabstack

Resources

Stars

4 stars

Watchers

0 watching

Forks

Releases

Packages

Used by

Contributors