# 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 Scraping API
  version: 1.0.0
extends: openapi/contextdev-web-scraping-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: 7
- target: $.paths['/web/scrape/html'].get
  update:
    x-apievangelist-phrasing:
      intent: Scrape a page's raw HTML
      effect: read
      questions:
      - Can I get the raw HTML of a web page, rendered in a browser?
      - Is there a way to keep only certain CSS selectors when scraping HTML?
      instructions:
      - text: Scrape the raw HTML of {url}.
        slots:
          url: query.url
      - text: Get the HTML of {url} keeping only elements matching {includeSelectors}.
        slots:
          url: query.url
          includeSelectors: query.includeSelectors
      method: generated
      generated: '2026-10-02'
- target: $.paths['/web/scrape/markdown'].get
  update:
    x-apievangelist-phrasing:
      intent: Scrape a single page into Markdown
      effect: read
      questions:
      - How do I turn one web page into clean Markdown for an LLM?
      - Can I strip headers, footers and navigation so only the main article comes through as Markdown?
      instructions:
      - text: Scrape {url} into Markdown.
        slots:
          url: query.url
      - text: Convert only the main content of {url} to Markdown {useMainContentOnly}.
        slots:
          url: query.url
          useMainContentOnly: query.useMainContentOnly
      method: generated
      generated: '2026-10-02'
- target: $.paths['/web/scrape/images'].get
  update:
    x-apievangelist-phrasing:
      intent: Collect all images on a web page
      effect: read
      questions:
      - Can I list every image on a page, including SVGs, CSS backgrounds and video posters?
      - Is there an option to remove visually duplicate images from the results?
      instructions:
      - text: Extract all images from {url}.
        slots:
          url: query.url
      - text: Get the images on {url} with duplicates removed {dedupe}.
        slots:
          url: query.url
          dedupe: query.dedupe
      method: generated
      generated: '2026-10-02'
- target: $.paths['/web/scrape/sitemap'].get
  update:
    x-apievangelist-phrasing:
      intent: List every page URL on a website
      effect: read
      questions:
      - How can I get a list of all the page URLs on a website from its sitemap?
      - Can I filter sitemap URLs with a regex, like only blog posts?
      instructions:
      - text: Crawl the sitemap of {domain} and list its URLs.
        slots:
          domain: query.domain
      - text: List up to {maxLinks} URLs on {domain} matching {urlRegex}.
        slots:
          maxLinks: query.maxLinks
          domain: query.domain
          urlRegex: query.urlRegex
      method: generated
      generated: '2026-10-02'
- target: $.paths['/web/crawl'].post
  update:
    x-apievangelist-phrasing:
      intent: Crawl a site and scrape many pages to Markdown
      effect: read
      questions:
      - Can I crawl a whole docs site from one starting URL and get every page back as Markdown?
      - What's the maximum number of pages a multi-page crawl will scrape?
      instructions:
      - text: Crawl starting at {url} and return every page as Markdown.
        slots:
          url: requestBody.url
      - text: Crawl {url} up to {maxPages} pages and depth {maxDepth}, scraping each to Markdown.
        slots:
          url: requestBody.url
          maxPages: requestBody.maxPages
          maxDepth: requestBody.maxDepth
      method: generated
      generated: '2026-10-02'
- target: $.paths['/web/search'].post
  update:
    x-apievangelist-phrasing:
      intent: Search the web and optionally scrape results
      effect: read
      questions:
      - Can I run a web search and get each result scraped into Markdown in one call?
      - Is it possible to limit web search results to certain domains or recent content?
      instructions:
      - text: Search the web for {query}.
        slots:
          query: requestBody.query
      - text: Search for {query} only on {includeDomains} published within {freshness}.
        slots:
          query: requestBody.query
          includeDomains: requestBody.includeDomains
          freshness: requestBody.freshness
      method: generated
      generated: '2026-10-02'
- target: $.paths['/web/screenshot'].get
  update:
    x-apievangelist-phrasing:
      intent: Capture a screenshot of a website
      effect: read
      questions:
      - Can I take a full-page screenshot of a website?
      - Does the screenshot tool dismiss cookie banners and support dark mode?
      instructions:
      - text: Take a screenshot of {domain}.
        slots:
          domain: query.domain
      - text: Capture a full-page screenshot of {directUrl} {fullScreenshot} in the {colorScheme} theme.
        slots:
          directUrl: query.directUrl
          fullScreenshot: query.fullScreenshot
          colorScheme: query.colorScheme
      method: generated
      generated: '2026-10-02'