API Evangelist conversational phrasing for Context Web Extraction API

# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand.
overlay: 1.0.0
info:
  title: API Evangelist conversational phrasing for Context Web Extraction API
  version: 1.0.0
extends: openapi/contextdev-web-extraction-api-openapi.yml
actions:
- target: $.info
  update:
    x-apievangelist-phrasing:
      method: generated
      generated: '2026-10-02'
      generator: build-phrasing.py
      label: Generated by API Evangelist
      operations: 9
- target: $.paths['/web/extract'].post
  update:
    x-apievangelist-phrasing:
      intent: Extract structured data from a site using a schema
      effect: read
      questions:
      - Can I crawl a website and get back data shaped to my own JSON Schema?
      - Is there a fact-check option so extracted values must be stated on the page?
      - How many pages will schema-based extraction analyze at most?
      instructions:
      - text: Extract data matching schema {schema} from the site at {url}.
        slots:
          schema: requestBody.schema
          url: requestBody.url
      - text: Crawl {url} up to {maxPages} pages and fill schema {schema} following {instructions}.
        slots:
          url: requestBody.url
          maxPages: requestBody.maxPages
          schema: requestBody.schema
          instructions: requestBody.instructions
      method: generated
      generated: '2026-10-02'
- target: $.paths['/web/competitors'].get
  update:
    x-apievangelist-phrasing:
      intent: Find a company's direct competitors
      effect: read
      questions:
      - Who are the direct competitors of a company, based on its website?
      - Can I choose exactly how many competitors come back for a domain?
      instructions:
      - text: Find direct competitors for {domain}.
        slots:
          domain: query.domain
      - text: List {numCompetitors} competitors of the company at {domain}.
        slots:
          numCompetitors: query.numCompetitors
          domain: query.domain
      method: generated
      generated: '2026-10-02'
- target: $.paths['/web/styleguide'].get
  update:
    x-apievangelist-phrasing:
      intent: Extract a website's design system
      effect: read
      questions:
      - Can I pull a site's full design system — colors, typography, spacing, shadows and components?
      - Does the styleguide extraction work against a site's dark mode?
      instructions:
      - text: Extract the styleguide from {domain}.
        slots:
          domain: query.domain
      - text: Get the design system of the page {directUrl} in the {colorScheme} color scheme.
        slots:
          directUrl: query.directUrl
          colorScheme: query.colorScheme
      method: generated
      generated: '2026-10-02'
- target: $.paths['/web/fonts'].get
  update:
    x-apievangelist-phrasing:
      intent: Find which fonts a website uses
      effect: read
      questions:
      - What font families does a website use, and how often is each one used?
      - Can I see the fallback fonts declared on a specific page?
      instructions:
      - text: List the fonts used on {domain}.
        slots:
          domain: query.domain
      - text: Scrape font families and usage counts from the page {directUrl}.
        slots:
          directUrl: query.directUrl
      method: generated
      generated: '2026-10-02'
- target: $.paths['/web/naics'].get
  update:
    x-apievangelist-phrasing:
      intent: Classify a brand into NAICS industry codes
      effect: read
      questions:
      - What NAICS industry code fits a company, given its domain or name?
      - Can I get up to ten 2022 NAICS codes ranked for a business?
      instructions:
      - text: Classify {input} into NAICS industry codes.
        slots:
          input: query.input
      - text: Return up to {maxResults} NAICS codes for the brand {input}.
        slots:
          maxResults: query.maxResults
          input: query.input
      method: generated
      generated: '2026-10-02'
- target: $.paths['/web/sic'].get
  update:
    x-apievangelist-phrasing:
      intent: Classify a brand into SIC industry codes
      effect: read
      questions:
      - Which SIC code does a company fall under, from its domain or name?
      - Can I classify against the SEC's current SIC list instead of the 1987 system?
      instructions:
      - text: Get SIC codes for {input}.
        slots:
          input: query.input
      - text: Classify {input} using the {type} SIC dataset.
        slots:
          input: query.input
          type: query.type
      method: generated
      generated: '2026-10-02'
- target: $.paths['/brand/ai/query'].post
  update:
    x-apievangelist-phrasing:
      intent: Ask AI to pull specific data points from a brand's site
      effect: read
      questions:
      - Can AI read a company's website and answer specific data points I list, like pricing or HQ address?
      - Is it possible to point the AI query at only certain pages of a brand's site?
      instructions:
      - text: Use AI to extract {data_to_extract} from {domain}'s website.
        slots:
          data_to_extract: requestBody.data_to_extract
          domain: requestBody.domain
      - text: Query {domain} for {data_to_extract}, looking only at pages {specific_pages}.
        slots:
          domain: requestBody.domain
          data_to_extract: requestBody.data_to_extract
          specific_pages: requestBody.specific_pages
      method: generated
      generated: '2026-10-02'
- target: $.paths['/brand/ai/product'].post
  update:
    x-apievangelist-phrasing:
      intent: Extract product details from one product page
      effect: read
      questions:
      - Given a single URL, can it tell me whether it's a product page and pull the product info?
      - What product details can I get from one specific product link?
      instructions:
      - text: Extract the product from the page {url}.
        slots:
          url: requestBody.url
      - text: Check whether {url} is a product page and return its product data.
        slots:
          url: requestBody.url
      method: generated
      generated: '2026-10-02'
- target: $.paths['/brand/ai/products'].post
  update:
    x-apievangelist-phrasing:
      intent: List the products a brand sells on its website
      effect: read
      questions:
      - Can I get a catalog of everything a brand sells, with pricing and features, from its website?
      - What details come back for each product when extracting a brand's whole product list?
      instructions:
      - text: Extract the full product list from the brand's website in the request.
      - text: Build a catalog of the brand's products with names, images and pricing.
      method: generated
      generated: '2026-10-02'