context.dev 1.17.0 → 1.19.0

data/rbi/context_dev/models/web_web_crawl_md_params.rbi

@@ -69,14 +69,13 @@ module ContextDev
       sig { params(max_pages: Integer).void }
       attr_writer :max_pages
 
-      # When true (default), PDF pages are fetched and their text layer is extracted and
-      # converted to Markdown alongside HTML pages. When false, PDF pages are skipped
-      # entirely (not included in results and not counted as failures).
-      sig { returns(T.nilable(T::Boolean)) }
-      attr_reader :parse_pdf
+      # PDF parsing controls. Use start/end to limit text extraction and OCR to an
+      # inclusive 1-based page range.
+      sig { returns(T.nilable(ContextDev::WebWebCrawlMdParams::Pdf)) }
+      attr_reader :pdf
 
-      sig { params(parse_pdf: T::Boolean).void }
-      attr_writer :parse_pdf
+      sig { params(pdf: ContextDev::WebWebCrawlMdParams::Pdf::OrHash).void }
+      attr_writer :pdf
 
       # Truncate base64-encoded image data in the Markdown output
       sig { returns(T.nilable(T::Boolean)) }
@@ -85,6 +84,16 @@ module ContextDev
       sig { params(shorten_base64_images: T::Boolean).void }
       attr_writer :shorten_base64_images
 
+      # Soft time budget for the crawl in milliseconds. After each scrape, the crawler
+      # checks the elapsed time and, if exceeded, returns the pages collected so far
+      # instead of continuing. Min: 10000 (10s). Max: 240000 (4 min). Default: 120000 (2
+      # min).
+      sig { returns(T.nilable(Integer)) }
+      attr_reader :stop_after_ms
+
+      sig { params(stop_after_ms: Integer).void }
+      attr_writer :stop_after_ms
+
       # Optional timeout in milliseconds for the request. If the request takes longer
       # than this value, it will be aborted with a 408 status code. Maximum allowed
       # value is 300000ms (5 minutes).
@@ -127,8 +136,9 @@ module ContextDev
           max_age_ms: Integer,
           max_depth: Integer,
           max_pages: Integer,
-          parse_pdf: T::Boolean,
+          pdf: ContextDev::WebWebCrawlMdParams::Pdf::OrHash,
           shorten_base64_images: T::Boolean,
+          stop_after_ms: Integer,
           timeout_ms: Integer,
           url_regex: String,
           use_main_content_only: T::Boolean,
@@ -158,12 +168,16 @@ module ContextDev
         max_depth: nil,
         # Maximum number of pages to crawl. Hard cap: 500.
         max_pages: nil,
-        # When true (default), PDF pages are fetched and their text layer is extracted and
-        # converted to Markdown alongside HTML pages. When false, PDF pages are skipped
-        # entirely (not included in results and not counted as failures).
-        parse_pdf: nil,
+        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
+        # inclusive 1-based page range.
+        pdf: nil,
         # Truncate base64-encoded image data in the Markdown output
         shorten_base64_images: nil,
+        # Soft time budget for the crawl in milliseconds. After each scrape, the crawler
+        # checks the elapsed time and, if exceeded, returns the pages collected so far
+        # instead of continuing. Min: 10000 (10s). Max: 240000 (4 min). Default: 120000 (2
+        # min).
+        stop_after_ms: nil,
         # Optional timeout in milliseconds for the request. If the request takes longer
         # than this value, it will be aborted with a 408 status code. Maximum allowed
         # value is 300000ms (5 minutes).
@@ -191,8 +205,9 @@ module ContextDev
             max_age_ms: Integer,
             max_depth: Integer,
             max_pages: Integer,
-            parse_pdf: T::Boolean,
+            pdf: ContextDev::WebWebCrawlMdParams::Pdf,
             shorten_base64_images: T::Boolean,
+            stop_after_ms: Integer,
             timeout_ms: Integer,
             url_regex: String,
             use_main_content_only: T::Boolean,
@@ -203,6 +218,68 @@ module ContextDev
       end
       def to_hash
       end
+
+      class Pdf < ContextDev::Internal::Type::BaseModel
+        OrHash =
+          T.type_alias do
+            T.any(
+              ContextDev::WebWebCrawlMdParams::Pdf,
+              ContextDev::Internal::AnyHash
+            )
+          end
+
+        # Last 1-based PDF page to parse. When omitted, parsing ends at the last page.
+        # Must be greater than or equal to start when both are provided.
+        sig { returns(T.nilable(Integer)) }
+        attr_reader :end_
+
+        sig { params(end_: Integer).void }
+        attr_writer :end_
+
+        # When true, PDF pages are fetched and parsed. When false, PDF pages are skipped
+        # entirely (not included in results and not counted as failures).
+        sig { returns(T.nilable(T::Boolean)) }
+        attr_reader :should_parse
+
+        sig { params(should_parse: T::Boolean).void }
+        attr_writer :should_parse
+
+        # First 1-based PDF page to parse. When omitted, parsing starts at the first page.
+        sig { returns(T.nilable(Integer)) }
+        attr_reader :start
+
+        sig { params(start: Integer).void }
+        attr_writer :start
+
+        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
+        # inclusive 1-based page range.
+        sig do
+          params(
+            end_: Integer,
+            should_parse: T::Boolean,
+            start: Integer
+          ).returns(T.attached_class)
+        end
+        def self.new(
+          # Last 1-based PDF page to parse. When omitted, parsing ends at the last page.
+          # Must be greater than or equal to start when both are provided.
+          end_: nil,
+          # When true, PDF pages are fetched and parsed. When false, PDF pages are skipped
+          # entirely (not included in results and not counted as failures).
+          should_parse: nil,
+          # First 1-based PDF page to parse. When omitted, parsing starts at the first page.
+          start: nil
+        )
+        end
+
+        sig do
+          override.returns(
+            { end_: Integer, should_parse: T::Boolean, start: Integer }
+          )
+        end
+        def to_hash
+        end
+      end
     end
   end
 end

data/rbi/context_dev/models/web_web_crawl_md_response.rbi

@@ -64,7 +64,8 @@ module ContextDev
         sig { returns(Integer) }
         attr_accessor :num_failed
 
-        # Number of URLs skipped (PDFs when parsePDF=false, or URLs not matching urlRegex)
+        # Number of URLs skipped (PDFs when pdf.shouldParse=false, or URLs not matching
+        # urlRegex)
         sig { returns(Integer) }
         attr_accessor :num_skipped
 
@@ -90,7 +91,8 @@ module ContextDev
           max_crawl_depth:,
           # Number of pages that failed to crawl
           num_failed:,
-          # Number of URLs skipped (PDFs when parsePDF=false, or URLs not matching urlRegex)
+          # Number of URLs skipped (PDFs when pdf.shouldParse=false, or URLs not matching
+          # urlRegex)
           num_skipped:,
           # Number of pages successfully crawled
           num_succeeded:,

data/rbi/context_dev/models/web_web_scrape_html_params.rbi

@@ -34,14 +34,13 @@ module ContextDev
       sig { params(max_age_ms: Integer).void }
       attr_writer :max_age_ms
 
-      # When true (default), PDF URLs are fetched and their text layer is extracted and
-      # returned wrapped in <html><pdf>…</pdf></html>. When false, PDF URLs are skipped
-      # and a 400 WEBSITE_ACCESS_ERROR is returned.
-      sig { returns(T.nilable(T::Boolean)) }
-      attr_reader :parse_pdf
+      # PDF parsing controls. Use start/end to limit text extraction and OCR to an
+      # inclusive 1-based page range.
+      sig { returns(T.nilable(ContextDev::WebWebScrapeHTMLParams::Pdf)) }
+      attr_reader :pdf
 
-      sig { params(parse_pdf: T::Boolean).void }
-      attr_writer :parse_pdf
+      sig { params(pdf: ContextDev::WebWebScrapeHTMLParams::Pdf::OrHash).void }
+      attr_writer :pdf
 
       # Optional timeout in milliseconds for the request. If the request takes longer
       # than this value, it will be aborted with a 408 status code. Maximum allowed
@@ -65,7 +64,7 @@ module ContextDev
           url: String,
           include_frames: T::Boolean,
           max_age_ms: Integer,
-          parse_pdf: T::Boolean,
+          pdf: ContextDev::WebWebScrapeHTMLParams::Pdf::OrHash,
           timeout_ms: Integer,
           wait_for_ms: Integer,
           request_options: ContextDev::RequestOptions::OrHash
@@ -80,10 +79,9 @@ module ContextDev
         # younger than this many milliseconds. Defaults to 1 day (86400000 ms) when
         # omitted. Max is 30 days (2592000000 ms). Set to 0 to always scrape fresh.
         max_age_ms: nil,
-        # When true (default), PDF URLs are fetched and their text layer is extracted and
-        # returned wrapped in <html><pdf>…</pdf></html>. When false, PDF URLs are skipped
-        # and a 400 WEBSITE_ACCESS_ERROR is returned.
-        parse_pdf: nil,
+        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
+        # inclusive 1-based page range.
+        pdf: nil,
         # Optional timeout in milliseconds for the request. If the request takes longer
         # than this value, it will be aborted with a 408 status code. Maximum allowed
         # value is 300000ms (5 minutes).
@@ -101,7 +99,7 @@ module ContextDev
             url: String,
             include_frames: T::Boolean,
             max_age_ms: Integer,
-            parse_pdf: T::Boolean,
+            pdf: ContextDev::WebWebScrapeHTMLParams::Pdf,
             timeout_ms: Integer,
             wait_for_ms: Integer,
             request_options: ContextDev::RequestOptions
@@ -110,6 +108,68 @@ module ContextDev
       end
       def to_hash
       end
+
+      class Pdf < ContextDev::Internal::Type::BaseModel
+        OrHash =
+          T.type_alias do
+            T.any(
+              ContextDev::WebWebScrapeHTMLParams::Pdf,
+              ContextDev::Internal::AnyHash
+            )
+          end
+
+        # Last 1-based PDF page to parse. When omitted, parsing ends at the last page.
+        # Must be greater than or equal to start when both are provided.
+        sig { returns(T.nilable(Integer)) }
+        attr_reader :end_
+
+        sig { params(end_: Integer).void }
+        attr_writer :end_
+
+        # When true, PDF URLs are fetched and parsed. When false, PDF URLs are skipped and
+        # a 400 WEBSITE_ACCESS_ERROR is returned.
+        sig { returns(T.nilable(T::Boolean)) }
+        attr_reader :should_parse
+
+        sig { params(should_parse: T::Boolean).void }
+        attr_writer :should_parse
+
+        # First 1-based PDF page to parse. When omitted, parsing starts at the first page.
+        sig { returns(T.nilable(Integer)) }
+        attr_reader :start
+
+        sig { params(start: Integer).void }
+        attr_writer :start
+
+        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
+        # inclusive 1-based page range.
+        sig do
+          params(
+            end_: Integer,
+            should_parse: T::Boolean,
+            start: Integer
+          ).returns(T.attached_class)
+        end
+        def self.new(
+          # Last 1-based PDF page to parse. When omitted, parsing ends at the last page.
+          # Must be greater than or equal to start when both are provided.
+          end_: nil,
+          # When true, PDF URLs are fetched and parsed. When false, PDF URLs are skipped and
+          # a 400 WEBSITE_ACCESS_ERROR is returned.
+          should_parse: nil,
+          # First 1-based PDF page to parse. When omitted, parsing starts at the first page.
+          start: nil
+        )
+        end
+
+        sig do
+          override.returns(
+            { end_: Integer, should_parse: T::Boolean, start: Integer }
+          )
+        end
+        def to_hash
+        end
+      end
     end
   end
 end

data/rbi/context_dev/models/web_web_scrape_md_params.rbi

@@ -46,14 +46,13 @@ module ContextDev
       sig { params(max_age_ms: Integer).void }
       attr_writer :max_age_ms
 
-      # When true (default), PDF URLs are fetched and their text layer is extracted and
-      # converted to Markdown. When false, PDF URLs are skipped and a 400
-      # WEBSITE_ACCESS_ERROR is returned.
-      sig { returns(T.nilable(T::Boolean)) }
-      attr_reader :parse_pdf
+      # PDF parsing controls. Use start/end to limit text extraction and OCR to an
+      # inclusive 1-based page range.
+      sig { returns(T.nilable(ContextDev::WebWebScrapeMdParams::Pdf)) }
+      attr_reader :pdf
 
-      sig { params(parse_pdf: T::Boolean).void }
-      attr_writer :parse_pdf
+      sig { params(pdf: ContextDev::WebWebScrapeMdParams::Pdf::OrHash).void }
+      attr_writer :pdf
 
       # Shorten base64-encoded image data in the Markdown output
       sig { returns(T.nilable(T::Boolean)) }
@@ -94,7 +93,7 @@ module ContextDev
           include_images: T::Boolean,
           include_links: T::Boolean,
           max_age_ms: Integer,
-          parse_pdf: T::Boolean,
+          pdf: ContextDev::WebWebScrapeMdParams::Pdf::OrHash,
           shorten_base64_images: T::Boolean,
           timeout_ms: Integer,
           use_main_content_only: T::Boolean,
@@ -116,10 +115,9 @@ module ContextDev
         # younger than this many milliseconds. Defaults to 1 day (86400000 ms) when
         # omitted. Max is 30 days (2592000000 ms). Set to 0 to always scrape fresh.
         max_age_ms: nil,
-        # When true (default), PDF URLs are fetched and their text layer is extracted and
-        # converted to Markdown. When false, PDF URLs are skipped and a 400
-        # WEBSITE_ACCESS_ERROR is returned.
-        parse_pdf: nil,
+        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
+        # inclusive 1-based page range.
+        pdf: nil,
         # Shorten base64-encoded image data in the Markdown output
         shorten_base64_images: nil,
         # Optional timeout in milliseconds for the request. If the request takes longer
@@ -144,7 +142,7 @@ module ContextDev
             include_images: T::Boolean,
             include_links: T::Boolean,
             max_age_ms: Integer,
-            parse_pdf: T::Boolean,
+            pdf: ContextDev::WebWebScrapeMdParams::Pdf,
             shorten_base64_images: T::Boolean,
             timeout_ms: Integer,
             use_main_content_only: T::Boolean,
@@ -155,6 +153,68 @@ module ContextDev
       end
       def to_hash
       end
+
+      class Pdf < ContextDev::Internal::Type::BaseModel
+        OrHash =
+          T.type_alias do
+            T.any(
+              ContextDev::WebWebScrapeMdParams::Pdf,
+              ContextDev::Internal::AnyHash
+            )
+          end
+
+        # Last 1-based PDF page to parse. When omitted, parsing ends at the last page.
+        # Must be greater than or equal to start when both are provided.
+        sig { returns(T.nilable(Integer)) }
+        attr_reader :end_
+
+        sig { params(end_: Integer).void }
+        attr_writer :end_
+
+        # When true, PDF URLs are fetched and parsed. When false, PDF URLs are skipped and
+        # a 400 WEBSITE_ACCESS_ERROR is returned.
+        sig { returns(T.nilable(T::Boolean)) }
+        attr_reader :should_parse
+
+        sig { params(should_parse: T::Boolean).void }
+        attr_writer :should_parse
+
+        # First 1-based PDF page to parse. When omitted, parsing starts at the first page.
+        sig { returns(T.nilable(Integer)) }
+        attr_reader :start
+
+        sig { params(start: Integer).void }
+        attr_writer :start
+
+        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
+        # inclusive 1-based page range.
+        sig do
+          params(
+            end_: Integer,
+            should_parse: T::Boolean,
+            start: Integer
+          ).returns(T.attached_class)
+        end
+        def self.new(
+          # Last 1-based PDF page to parse. When omitted, parsing ends at the last page.
+          # Must be greater than or equal to start when both are provided.
+          end_: nil,
+          # When true, PDF URLs are fetched and parsed. When false, PDF URLs are skipped and
+          # a 400 WEBSITE_ACCESS_ERROR is returned.
+          should_parse: nil,
+          # First 1-based PDF page to parse. When omitted, parsing starts at the first page.
+          start: nil
+        )
+        end
+
+        sig do
+          override.returns(
+            { end_: Integer, should_parse: T::Boolean, start: Integer }
+          )
+        end
+        def to_hash
+        end
+      end
     end
   end
 end

data/rbi/context_dev/resources/web.rbi

@@ -65,6 +65,8 @@ module ContextDev
           domain: String,
           full_screenshot:
             ContextDev::WebScreenshotParams::FullScreenshot::OrSymbol,
+          handle_cookie_popup:
+            ContextDev::WebScreenshotParams::HandleCookiePopup::OrSymbol,
           max_age_ms: Integer,
           page: ContextDev::WebScreenshotParams::Page::OrSymbol,
           timeout_ms: Integer,
@@ -86,6 +88,10 @@ module ContextDev
         # screenshot capturing all content. If 'false' or not provided, takes a viewport
         # screenshot (standard browser view).
         full_screenshot: nil,
+        # Optional parameter to control cookie/consent popup handling. If 'true', we
+        # dismiss cookie banner before capture. If 'false' or not provided, captures the
+        # page without that step.
+        handle_cookie_popup: nil,
         # Return a cached screenshot if a prior screenshot for the same parameters exists
         # and is younger than this many milliseconds. Defaults to 1 day (86400000 ms) when
         # omitted. Max is 30 days (2592000000 ms). Set to 0 to always capture fresh.
@@ -122,8 +128,9 @@ module ContextDev
           max_age_ms: Integer,
           max_depth: Integer,
           max_pages: Integer,
-          parse_pdf: T::Boolean,
+          pdf: ContextDev::WebWebCrawlMdParams::Pdf::OrHash,
           shorten_base64_images: T::Boolean,
+          stop_after_ms: Integer,
           timeout_ms: Integer,
           url_regex: String,
           use_main_content_only: T::Boolean,
@@ -153,12 +160,16 @@ module ContextDev
         max_depth: nil,
         # Maximum number of pages to crawl. Hard cap: 500.
         max_pages: nil,
-        # When true (default), PDF pages are fetched and their text layer is extracted and
-        # converted to Markdown alongside HTML pages. When false, PDF pages are skipped
-        # entirely (not included in results and not counted as failures).
-        parse_pdf: nil,
+        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
+        # inclusive 1-based page range.
+        pdf: nil,
         # Truncate base64-encoded image data in the Markdown output
         shorten_base64_images: nil,
+        # Soft time budget for the crawl in milliseconds. After each scrape, the crawler
+        # checks the elapsed time and, if exceeded, returns the pages collected so far
+        # instead of continuing. Min: 10000 (10s). Max: 240000 (4 min). Default: 120000 (2
+        # min).
+        stop_after_ms: nil,
         # Optional timeout in milliseconds for the request. If the request takes longer
         # than this value, it will be aborted with a 408 status code. Maximum allowed
         # value is 300000ms (5 minutes).
@@ -181,7 +192,7 @@ module ContextDev
           url: String,
           include_frames: T::Boolean,
           max_age_ms: Integer,
-          parse_pdf: T::Boolean,
+          pdf: ContextDev::WebWebScrapeHTMLParams::Pdf::OrHash,
           timeout_ms: Integer,
           wait_for_ms: Integer,
           request_options: ContextDev::RequestOptions::OrHash
@@ -196,10 +207,9 @@ module ContextDev
         # younger than this many milliseconds. Defaults to 1 day (86400000 ms) when
         # omitted. Max is 30 days (2592000000 ms). Set to 0 to always scrape fresh.
         max_age_ms: nil,
-        # When true (default), PDF URLs are fetched and their text layer is extracted and
-        # returned wrapped in <html><pdf>…</pdf></html>. When false, PDF URLs are skipped
-        # and a 400 WEBSITE_ACCESS_ERROR is returned.
-        parse_pdf: nil,
+        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
+        # inclusive 1-based page range.
+        pdf: nil,
         # Optional timeout in milliseconds for the request. If the request takes longer
         # than this value, it will be aborted with a 408 status code. Maximum allowed
         # value is 300000ms (5 minutes).
@@ -253,7 +263,7 @@ module ContextDev
           include_images: T::Boolean,
           include_links: T::Boolean,
           max_age_ms: Integer,
-          parse_pdf: T::Boolean,
+          pdf: ContextDev::WebWebScrapeMdParams::Pdf::OrHash,
           shorten_base64_images: T::Boolean,
           timeout_ms: Integer,
           use_main_content_only: T::Boolean,
@@ -275,10 +285,9 @@ module ContextDev
         # younger than this many milliseconds. Defaults to 1 day (86400000 ms) when
         # omitted. Max is 30 days (2592000000 ms). Set to 0 to always scrape fresh.
         max_age_ms: nil,
-        # When true (default), PDF URLs are fetched and their text layer is extracted and
-        # converted to Markdown. When false, PDF URLs are skipped and a 400
-        # WEBSITE_ACCESS_ERROR is returned.
-        parse_pdf: nil,
+        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
+        # inclusive 1-based page range.
+        pdf: nil,
         # Shorten base64-encoded image data in the Markdown output
         shorten_base64_images: nil,
         # Optional timeout in milliseconds for the request. If the request takes longer

data/sig/context_dev/models/web_screenshot_params.rbs

@@ -5,6 +5,7 @@ module ContextDev
         direct_url: String,
         domain: String,
         full_screenshot: ContextDev::Models::WebScreenshotParams::full_screenshot,
+        handle_cookie_popup: ContextDev::Models::WebScreenshotParams::handle_cookie_popup,
         max_age_ms: Integer,
         page: ContextDev::Models::WebScreenshotParams::page,
         timeout_ms: Integer,
@@ -31,6 +32,12 @@ module ContextDev
         ContextDev::Models::WebScreenshotParams::full_screenshot
       ) -> ContextDev::Models::WebScreenshotParams::full_screenshot
 
+      attr_reader handle_cookie_popup: ContextDev::Models::WebScreenshotParams::handle_cookie_popup?
+
+      def handle_cookie_popup=: (
+        ContextDev::Models::WebScreenshotParams::handle_cookie_popup
+      ) -> ContextDev::Models::WebScreenshotParams::handle_cookie_popup
+
       attr_reader max_age_ms: Integer?
 
       def max_age_ms=: (Integer) -> Integer
@@ -59,6 +66,7 @@ module ContextDev
         ?direct_url: String,
         ?domain: String,
         ?full_screenshot: ContextDev::Models::WebScreenshotParams::full_screenshot,
+        ?handle_cookie_popup: ContextDev::Models::WebScreenshotParams::handle_cookie_popup,
         ?max_age_ms: Integer,
         ?page: ContextDev::Models::WebScreenshotParams::page,
         ?timeout_ms: Integer,
@@ -71,6 +79,7 @@ module ContextDev
         direct_url: String,
         domain: String,
         full_screenshot: ContextDev::Models::WebScreenshotParams::full_screenshot,
+        handle_cookie_popup: ContextDev::Models::WebScreenshotParams::handle_cookie_popup,
         max_age_ms: Integer,
         page: ContextDev::Models::WebScreenshotParams::page,
         timeout_ms: Integer,
@@ -90,6 +99,17 @@ module ContextDev
         def self?.values: -> ::Array[ContextDev::Models::WebScreenshotParams::full_screenshot]
       end
 
+      type handle_cookie_popup = :true | :false
+
+      module HandleCookiePopup
+        extend ContextDev::Internal::Type::Enum
+
+        TRUE: :true
+        FALSE: :false
+
+        def self?.values: -> ::Array[ContextDev::Models::WebScreenshotParams::handle_cookie_popup]
+      end
+
       type page =
         :login
         | :signup

data/sig/context_dev/models/web_web_crawl_md_params.rbs

@@ -10,8 +10,9 @@ module ContextDev
         max_age_ms: Integer,
         max_depth: Integer,
         max_pages: Integer,
-        parse_pdf: bool,
+        pdf: ContextDev::WebWebCrawlMdParams::Pdf,
         :shorten_base64_images => bool,
+        stop_after_ms: Integer,
         timeout_ms: Integer,
         url_regex: String,
         use_main_content_only: bool,
@@ -53,14 +54,20 @@ module ContextDev
 
       def max_pages=: (Integer) -> Integer
 
-      attr_reader parse_pdf: bool?
+      attr_reader pdf: ContextDev::WebWebCrawlMdParams::Pdf?
 
-      def parse_pdf=: (bool) -> bool
+      def pdf=: (
+        ContextDev::WebWebCrawlMdParams::Pdf
+      ) -> ContextDev::WebWebCrawlMdParams::Pdf
 
       attr_reader shorten_base64_images: bool?
 
       def shorten_base64_images=: (bool) -> bool
 
+      attr_reader stop_after_ms: Integer?
+
+      def stop_after_ms=: (Integer) -> Integer
+
       attr_reader timeout_ms: Integer?
 
       def timeout_ms=: (Integer) -> Integer
@@ -86,8 +93,9 @@ module ContextDev
         ?max_age_ms: Integer,
         ?max_depth: Integer,
         ?max_pages: Integer,
-        ?parse_pdf: bool,
+        ?pdf: ContextDev::WebWebCrawlMdParams::Pdf,
         ?shorten_base64_images: bool,
+        ?stop_after_ms: Integer,
         ?timeout_ms: Integer,
         ?url_regex: String,
         ?use_main_content_only: bool,
@@ -104,14 +112,39 @@ module ContextDev
         max_age_ms: Integer,
         max_depth: Integer,
         max_pages: Integer,
-        parse_pdf: bool,
+        pdf: ContextDev::WebWebCrawlMdParams::Pdf,
         :shorten_base64_images => bool,
+        stop_after_ms: Integer,
         timeout_ms: Integer,
         url_regex: String,
         use_main_content_only: bool,
         wait_for_ms: Integer,
         request_options: ContextDev::RequestOptions
       }
+
+      type pdf = { end_: Integer, should_parse: bool, start: Integer }
+
+      class Pdf < ContextDev::Internal::Type::BaseModel
+        attr_reader end_: Integer?
+
+        def end_=: (Integer) -> Integer
+
+        attr_reader should_parse: bool?
+
+        def should_parse=: (bool) -> bool
+
+        attr_reader start: Integer?
+
+        def start=: (Integer) -> Integer
+
+        def initialize: (
+          ?end_: Integer,
+          ?should_parse: bool,
+          ?start: Integer
+        ) -> void
+
+        def to_hash: -> { end_: Integer, should_parse: bool, start: Integer }
+      end
     end
   end
 end