NEVER TRAINED ON · PRIVATE BY DEFAULT · BUILT BY MOZILLA

GITHUB

Schema library b2b-intel

Return a press release or news mention from a public page.

Pass this JSON Schema and a public source URL to/extract/json. Normalize public announcements and coverage into a record for research or monitoring.

tabstack.ai/pricing

TEXT

MARKDOWN

LINKS

extract.tsTypeScript
const res = await client.extract.json({
  url: 'https://example.com/pricing',
  json_schema: {
    type: 'object',
    properties: {
      plan: { type: 'string' },
      price: { type: 'number' },
    },
  },
})

What it captures

Fields in this press release or news mention.

Schema for press releases and news mentions about B2B companies.

FieldTypeWhat it holds
headlinestring, requiredHeadline or title of the press release or news article.
company_namestring, requiredPrimary company mentioned or featured.
company_domainstringWebsite domain of the primary company.
publication_namestringName of the publication or newswire.
authorstringAuthor of the article or press release.
publish_datestring, requiredDate the article or press release was published.
mention_typestring, requiredCategory of the news mention.
summarystringShort summary or lead paragraph.
full_textstringFull text content of the press release or article.
sentimentstringInferred sentiment of the mention.
quote_from_execstringNotable executive quote from the article.
related_companiesarrayOther companies mentioned in the article.
related_productsarrayProducts mentioned in the article.
page_titlestringTitle of the source page. Tabstack auto-fills this from page metadata when left empty.
faviconstringFavicon URL of the source page. Tabstack auto-fills this from page metadata when left empty.

Example

Inspect and validate the response shape.

The example below is generated from the schema to demonstrate its structure. Edit the object or paste a real response to validate it in your browser.

What it checks

Pass or fail, and the first reason why.

Whether the text parses as JSON, whether the top level is an object, and whether each field the schema requires is present at the declared type. A null is allowed anywhere. Every key you ask for is present. A field the page does not state can come back null, empty, a placeholder number, or a guessed value. Validate values, not just keys.

Example data, not a live response.

Usage

Send the schema with your source URL.

The schema travels with the request rather than living on your account, so the same call can send a trimmed version for one page and the full one for another. Every key you ask for is present. A field the page does not state can come back null, empty, a placeholder number, or a guessed value. Validate values, not just keys.

press.tsTypeScript
import schema from './press-release-news-mention.json'

const res = await client.extract.json({
  url: 'https://example.com/pricing',
  json_schema: schema,
})

Adaptation

Make the schema match your application.

The schema is a starting point, not a guarantee that every source page contains every field. Three things are worth doing before you write one into your system, and the schema-authoring guide covers the rest.

Remove what you do not need

A shorter schema is a smaller response and fewer fields to handle.

Describe the ambiguous ones

A description tells Extract what to look for when a label could mean two things.

Validate before you store

Check the returned object against the schema rather than trusting it.

Related

Related schemas

Other schemas in this category.

START FREE

Try this schema on a public page.

Start with 10,000 free credits. No credit card required.

curl -X POST https://api.tabstack.ai/v1/extract/json -H "Authorization: Bearer $TABSTACK_API_KEY" -d '{"url":"...","json_schema":{...}}'