context.dev 1.4.0 → 1.6.0

data/rbi/context_dev/models/web_screenshot_params.rbi

@@ -13,7 +13,7 @@ module ContextDev
 
       # A specific URL to screenshot directly, bypassing domain resolution (e.g.,
       # 'https://example.com/pricing'). When provided, the screenshot is taken of this
-      # exact URL.
+      # exact URL. You must provide either 'domain' or 'directUrl', but not both.
       sig { returns(T.nilable(String)) }
       attr_reader :direct_url
 
@@ -21,7 +21,8 @@ module ContextDev
       attr_writer :direct_url
 
       # Domain name to take screenshot of (e.g., 'example.com', 'google.com'). The
-      # domain will be automatically normalized and validated.
+      # domain will be automatically normalized and validated. You must provide either
+      # 'domain' or 'directUrl', but not both.
       sig { returns(T.nilable(String)) }
       attr_reader :domain
 
@@ -90,10 +91,11 @@ module ContextDev
       def self.new(
         # A specific URL to screenshot directly, bypassing domain resolution (e.g.,
         # 'https://example.com/pricing'). When provided, the screenshot is taken of this
-        # exact URL.
+        # exact URL. You must provide either 'domain' or 'directUrl', but not both.
         direct_url: nil,
         # Domain name to take screenshot of (e.g., 'example.com', 'google.com'). The
-        # domain will be automatically normalized and validated.
+        # domain will be automatically normalized and validated. You must provide either
+        # 'domain' or 'directUrl', but not both.
         domain: nil,
         # Optional parameter to determine screenshot type. If 'true', takes a full page
         # screenshot capturing all content. If 'false' or not provided, takes a viewport

data/rbi/context_dev/models/web_web_crawl_md_params.rbi

@@ -38,6 +38,15 @@ module ContextDev
       sig { params(include_links: T::Boolean).void }
       attr_writer :include_links
 
+      # Return a cached result if a prior scrape 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 scrape fresh.
+      sig { returns(T.nilable(Integer)) }
+      attr_reader :max_age_ms
+
+      sig { params(max_age_ms: Integer).void }
+      attr_writer :max_age_ms
+
       # Maximum link depth from the starting URL (0 = only the starting page)
       sig { returns(T.nilable(Integer)) }
       attr_reader :max_depth
@@ -80,6 +89,7 @@ module ContextDev
           follow_subdomains: T::Boolean,
           include_images: T::Boolean,
           include_links: T::Boolean,
+          max_age_ms: Integer,
           max_depth: Integer,
           max_pages: Integer,
           shorten_base64_images: T::Boolean,
@@ -99,6 +109,10 @@ module ContextDev
         include_images: nil,
         # Preserve hyperlinks in the Markdown output
         include_links: nil,
+        # Return a cached result if a prior scrape 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 scrape fresh.
+        max_age_ms: nil,
         # Maximum link depth from the starting URL (0 = only the starting page)
         max_depth: nil,
         # Maximum number of pages to crawl. Hard cap: 500.
@@ -121,6 +135,7 @@ module ContextDev
             follow_subdomains: T::Boolean,
             include_images: T::Boolean,
             include_links: T::Boolean,
+            max_age_ms: Integer,
             max_depth: Integer,
             max_pages: Integer,
             shorten_base64_images: T::Boolean,

data/rbi/context_dev/models/web_web_scrape_html_params.rbi

@@ -18,22 +18,40 @@ module ContextDev
       sig { returns(String) }
       attr_accessor :url
 
+      # Return a cached result if a prior scrape 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 scrape fresh.
+      sig { returns(T.nilable(Integer)) }
+      attr_reader :max_age_ms
+
+      sig { params(max_age_ms: Integer).void }
+      attr_writer :max_age_ms
+
       sig do
         params(
           url: String,
+          max_age_ms: Integer,
           request_options: ContextDev::RequestOptions::OrHash
         ).returns(T.attached_class)
       end
       def self.new(
         # Full URL to scrape (must include http:// or https:// protocol)
         url:,
+        # Return a cached result if a prior scrape 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 scrape fresh.
+        max_age_ms: nil,
         request_options: {}
       )
       end
 
       sig do
         override.returns(
-          { url: String, request_options: ContextDev::RequestOptions }
+          {
+            url: String,
+            max_age_ms: Integer,
+            request_options: ContextDev::RequestOptions
+          }
         )
       end
       def to_hash

data/rbi/context_dev/models/web_web_scrape_md_params.rbi

@@ -11,7 +11,7 @@ module ContextDev
           T.any(ContextDev::WebWebScrapeMdParams, ContextDev::Internal::AnyHash)
         end
 
-      # Full URL to scrape and convert to markdown (must include http:// or https://
+      # Full URL to scrape into LLM usable Markdown (must include http:// or https://
       # protocol)
       sig { returns(String) }
       attr_accessor :url
@@ -30,6 +30,15 @@ module ContextDev
       sig { params(include_links: T::Boolean).void }
       attr_writer :include_links
 
+      # Return a cached result if a prior scrape 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 scrape fresh.
+      sig { returns(T.nilable(Integer)) }
+      attr_reader :max_age_ms
+
+      sig { params(max_age_ms: Integer).void }
+      attr_writer :max_age_ms
+
       # Shorten base64-encoded image data in the Markdown output
       sig { returns(T.nilable(T::Boolean)) }
       attr_reader :shorten_base64_images
@@ -50,19 +59,24 @@ module ContextDev
           url: String,
           include_images: T::Boolean,
           include_links: T::Boolean,
+          max_age_ms: Integer,
           shorten_base64_images: T::Boolean,
           use_main_content_only: T::Boolean,
           request_options: ContextDev::RequestOptions::OrHash
         ).returns(T.attached_class)
       end
       def self.new(
-        # Full URL to scrape and convert to markdown (must include http:// or https://
+        # Full URL to scrape into LLM usable Markdown (must include http:// or https://
         # protocol)
         url:,
         # Include image references in Markdown output
         include_images: nil,
         # Preserve hyperlinks in Markdown output
         include_links: nil,
+        # Return a cached result if a prior scrape 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 scrape fresh.
+        max_age_ms: nil,
         # Shorten base64-encoded image data in the Markdown output
         shorten_base64_images: nil,
         # Extract only the main content of the page, excluding headers, footers, sidebars,
@@ -78,6 +92,7 @@ module ContextDev
             url: String,
             include_images: T::Boolean,
             include_links: T::Boolean,
+            max_age_ms: Integer,
             shorten_base64_images: T::Boolean,
             use_main_content_only: T::Boolean,
             request_options: ContextDev::RequestOptions

data/rbi/context_dev/models/web_web_scrape_sitemap_params.rbi

@@ -14,8 +14,7 @@ module ContextDev
           )
         end
 
-      # Domain name to crawl sitemaps for (e.g., 'example.com'). The domain will be
-      # automatically normalized and validated.
+      # Domain to build a sitemap for
       sig { returns(String) }
       attr_accessor :domain
 
@@ -27,20 +26,31 @@ module ContextDev
       sig { params(max_links: Integer).void }
       attr_writer :max_links
 
+      # Optional RE2-compatible regex pattern. Only URLs matching this pattern are
+      # returned and counted against maxLinks.
+      sig { returns(T.nilable(String)) }
+      attr_reader :url_regex
+
+      sig { params(url_regex: String).void }
+      attr_writer :url_regex
+
       sig do
         params(
           domain: String,
           max_links: Integer,
+          url_regex: String,
           request_options: ContextDev::RequestOptions::OrHash
         ).returns(T.attached_class)
       end
       def self.new(
-        # Domain name to crawl sitemaps for (e.g., 'example.com'). The domain will be
-        # automatically normalized and validated.
+        # Domain to build a sitemap for
         domain:,
         # Maximum number of links to return from the sitemap crawl. Defaults to 10,000.
         # Minimum is 1, maximum is 100,000.
         max_links: nil,
+        # Optional RE2-compatible regex pattern. Only URLs matching this pattern are
+        # returned and counted against maxLinks.
+        url_regex: nil,
         request_options: {}
       )
       end
@@ -50,6 +60,7 @@ module ContextDev
           {
             domain: String,
             max_links: Integer,
+            url_regex: String,
             request_options: ContextDev::RequestOptions
           }
         )

data/rbi/context_dev/models.rbi

@@ -25,16 +25,15 @@ module ContextDev
 
   IndustryRetrieveNaicsParams = ContextDev::Models::IndustryRetrieveNaicsParams
 
-  StyleExtractFontsParams = ContextDev::Models::StyleExtractFontsParams
-
-  StyleExtractStyleguideParams =
-    ContextDev::Models::StyleExtractStyleguideParams
-
   UtilityPrefetchByEmailParams =
     ContextDev::Models::UtilityPrefetchByEmailParams
 
   UtilityPrefetchParams = ContextDev::Models::UtilityPrefetchParams
 
+  WebExtractFontsParams = ContextDev::Models::WebExtractFontsParams
+
+  WebExtractStyleguideParams = ContextDev::Models::WebExtractStyleguideParams
+
   WebScreenshotParams = ContextDev::Models::WebScreenshotParams
 
   WebWebCrawlMdParams = ContextDev::Models::WebWebCrawlMdParams

data/rbi/context_dev/resources/brand.rbi

@@ -63,7 +63,6 @@ module ContextDev
         force_language: nil,
         # When set to true, the API will perform an additional verification steps to
         # ensure the identified brand matches the transaction with high confidence.
-        # Defaults to false.
         high_confidence_only: nil,
         # Optional parameter to optimize the API call for maximum speed. When set to true,
         # the API will skip time-consuming operations for faster response at the cost of
@@ -83,9 +82,8 @@ module ContextDev
       end
 
       # Retrieve brand information using an email address while detecting disposable and
-      # free email addresses. This endpoint extracts the domain from the email address
-      # and returns brand data for that domain. Disposable and free email addresses
-      # (like gmail.com, yahoo.com) will throw a 422 error.
+      # free email addresses. Disposable and free email addresses (like gmail.com,
+      # yahoo.com) will throw a 422 error.
       sig do
         params(
           email: String,
@@ -116,8 +114,7 @@ module ContextDev
       end
 
       # Retrieve brand information using an ISIN (International Securities
-      # Identification Number). This endpoint looks up the company associated with the
-      # ISIN and returns its brand data.
+      # Identification Number).
       sig do
         params(
           isin: String,
@@ -147,8 +144,7 @@ module ContextDev
       )
       end
 
-      # Retrieve brand information using a company name. This endpoint searches for the
-      # company by name and returns its brand data.
+      # Retrieve brand information using a company name.
       sig do
         params(
           name: String,
@@ -165,8 +161,8 @@ module ContextDev
         # Company name to retrieve brand data for (e.g., 'Apple Inc', 'Microsoft
         # Corporation'). Must be 3-30 characters.
         name:,
-        # Optional country code (GL parameter) to specify the country. This affects the
-        # geographic location used for search queries.
+        # Optional country code hint (GL parameter) to specify the country for the company
+        # name.
         country_gl: nil,
         # Optional parameter to force the language of the retrieved brand data.
         force_language: nil,
@@ -182,8 +178,7 @@ module ContextDev
       )
       end
 
-      # Retrieve brand information using a stock ticker symbol. This endpoint looks up
-      # the company associated with the ticker and returns its brand data.
+      # Retrieve brand information using a stock ticker symbol.
       sig do
         params(
           ticker: String,
@@ -217,8 +212,8 @@ module ContextDev
       end
 
       # Returns a simplified version of brand data containing only essential
-      # information: domain, title, colors, logos, and backdrops. This endpoint is
-      # optimized for faster responses and reduced data transfer.
+      # information: domain, title, colors, logos, and backdrops. Optimized for faster
+      # responses and reduced data transfer.
       sig do
         params(
           domain: String,

data/rbi/context_dev/resources/industry.rbi

@@ -3,7 +3,7 @@
 module ContextDev
   module Resources
     class Industry
-      # Endpoint to classify any brand into a 2022 NAICS code.
+      # Classify any brand into 2022 NAICS industry codes from its domain or name.
       sig do
         params(
           input: String,
@@ -14,9 +14,9 @@ module ContextDev
         ).returns(ContextDev::Models::IndustryRetrieveNaicsResponse)
       end
       def retrieve_naics(
-        # Brand domain or title to retrieve NAICS code for. If a valid domain is provided
-        # in `input`, it will be used for classification, otherwise, we will search for
-        # the brand using the provided title.
+        # Brand domain or title to retrieve NAICS code for. If a valid domain is provided,
+        # it will be used for classification, otherwise, we will search for the brand
+        # using the provided title.
         input:,
         # Maximum number of NAICS codes to return. Must be between 1 and 10. Defaults
         # to 5.

data/rbi/context_dev/resources/web.rbi

@@ -3,11 +3,62 @@
 module ContextDev
   module Resources
     class Web
-      # Capture a screenshot of a website. Supports both viewport (standard browser
-      # view) and full-page screenshots. Can also screenshot specific page types (login,
-      # pricing, etc.) by using heuristics to find the appropriate URL. Either 'domain'
-      # or 'directUrl' must be provided as a query parameter, but not both. Returns a
-      # URL to the uploaded screenshot image hosted on our CDN.
+      # Scrape font information from a website including font families, usage
+      # statistics, fallbacks, and element/word counts.
+      sig do
+        params(
+          direct_url: String,
+          domain: String,
+          timeout_ms: Integer,
+          request_options: ContextDev::RequestOptions::OrHash
+        ).returns(ContextDev::Models::WebExtractFontsResponse)
+      end
+      def extract_fonts(
+        # A specific URL to fetch fonts from directly, bypassing domain resolution (e.g.,
+        # 'https://example.com/design-system'). When provided, fonts are extracted from
+        # this exact URL. You must provide either 'domain' or 'directUrl', but not both.
+        direct_url: nil,
+        # Domain name to extract fonts from (e.g., 'example.com', 'google.com'). The
+        # domain will be automatically normalized and validated. You must provide either
+        # 'domain' or 'directUrl', but not both.
+        domain: 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).
+        timeout_ms: nil,
+        request_options: {}
+      )
+      end
+
+      # Extract a comprehensive design system from a website including colors,
+      # typography, spacing, shadows, and UI components.
+      sig do
+        params(
+          direct_url: String,
+          domain: String,
+          timeout_ms: Integer,
+          request_options: ContextDev::RequestOptions::OrHash
+        ).returns(ContextDev::Models::WebExtractStyleguideResponse)
+      end
+      def extract_styleguide(
+        # A specific URL to fetch the styleguide from directly, bypassing domain
+        # resolution (e.g., 'https://example.com/design-system'). When provided, the
+        # styleguide is extracted from this exact URL. You must provide either 'domain' or
+        # 'directUrl', but not both.
+        direct_url: nil,
+        # Domain name to extract styleguide from (e.g., 'example.com', 'google.com'). The
+        # domain will be automatically normalized and validated. You must provide either
+        # 'domain' or 'directUrl', but not both.
+        domain: 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).
+        timeout_ms: nil,
+        request_options: {}
+      )
+      end
+
+      # Capture a screenshot of a website.
       sig do
         params(
           direct_url: String,
@@ -22,10 +73,11 @@ module ContextDev
       def screenshot(
         # A specific URL to screenshot directly, bypassing domain resolution (e.g.,
         # 'https://example.com/pricing'). When provided, the screenshot is taken of this
-        # exact URL.
+        # exact URL. You must provide either 'domain' or 'directUrl', but not both.
         direct_url: nil,
         # Domain name to take screenshot of (e.g., 'example.com', 'google.com'). The
-        # domain will be automatically normalized and validated.
+        # domain will be automatically normalized and validated. You must provide either
+        # 'domain' or 'directUrl', but not both.
         domain: nil,
         # Optional parameter to determine screenshot type. If 'true', takes a full page
         # screenshot capturing all content. If 'false' or not provided, takes a viewport
@@ -46,14 +98,14 @@ module ContextDev
       end
 
       # Performs a crawl starting from a given URL, extracts page content as Markdown,
-      # and returns results for all crawled pages. Only follows links within the same
-      # domain as the starting URL. Costs 1 credit per successful page crawled.
+      # and returns results for all crawled pages.
       sig do
         params(
           url: String,
           follow_subdomains: T::Boolean,
           include_images: T::Boolean,
           include_links: T::Boolean,
+          max_age_ms: Integer,
           max_depth: Integer,
           max_pages: Integer,
           shorten_base64_images: T::Boolean,
@@ -73,6 +125,10 @@ module ContextDev
         include_images: nil,
         # Preserve hyperlinks in the Markdown output
         include_links: nil,
+        # Return a cached result if a prior scrape 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 scrape fresh.
+        max_age_ms: nil,
         # Maximum link depth from the starting URL (0 = only the starting page)
         max_depth: nil,
         # Maximum number of pages to crawl. Hard cap: 500.
@@ -92,12 +148,17 @@ module ContextDev
       sig do
         params(
           url: String,
+          max_age_ms: Integer,
           request_options: ContextDev::RequestOptions::OrHash
         ).returns(ContextDev::Models::WebWebScrapeHTMLResponse)
       end
       def web_scrape_html(
         # Full URL to scrape (must include http:// or https:// protocol)
         url:,
+        # Return a cached result if a prior scrape 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 scrape fresh.
+        max_age_ms: nil,
         request_options: {}
       )
       end
@@ -118,26 +179,30 @@ module ContextDev
       )
       end
 
-      # Scrapes the given URL, converts the HTML content to Markdown, and returns the
-      # result.
+      # Scrapes the given URL into LLM usable Markdown.
       sig do
         params(
           url: String,
           include_images: T::Boolean,
           include_links: T::Boolean,
+          max_age_ms: Integer,
           shorten_base64_images: T::Boolean,
           use_main_content_only: T::Boolean,
           request_options: ContextDev::RequestOptions::OrHash
         ).returns(ContextDev::Models::WebWebScrapeMdResponse)
       end
       def web_scrape_md(
-        # Full URL to scrape and convert to markdown (must include http:// or https://
+        # Full URL to scrape into LLM usable Markdown (must include http:// or https://
         # protocol)
         url:,
         # Include image references in Markdown output
         include_images: nil,
         # Preserve hyperlinks in Markdown output
         include_links: nil,
+        # Return a cached result if a prior scrape 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 scrape fresh.
+        max_age_ms: nil,
         # Shorten base64-encoded image data in the Markdown output
         shorten_base64_images: nil,
         # Extract only the main content of the page, excluding headers, footers, sidebars,
@@ -147,23 +212,24 @@ module ContextDev
       )
       end
 
-      # Crawls the sitemap of the given domain and returns all discovered page URLs.
-      # Supports sitemap index files (recursive), parallel fetching with concurrency
-      # control, deduplication, and filters out non-page resources (images, PDFs, etc.).
+      # Crawl an entire website's sitemap and return all discovered page URLs.
       sig do
         params(
           domain: String,
           max_links: Integer,
+          url_regex: String,
           request_options: ContextDev::RequestOptions::OrHash
         ).returns(ContextDev::Models::WebWebScrapeSitemapResponse)
       end
       def web_scrape_sitemap(
-        # Domain name to crawl sitemaps for (e.g., 'example.com'). The domain will be
-        # automatically normalized and validated.
+        # Domain to build a sitemap for
         domain:,
         # Maximum number of links to return from the sitemap crawl. Defaults to 10,000.
         # Minimum is 1, maximum is 100,000.
         max_links: nil,
+        # Optional RE2-compatible regex pattern. Only URLs matching this pattern are
+        # returned and counted against maxLinks.
+        url_regex: nil,
         request_options: {}
       )
       end

data/sig/context_dev/client.rbs

@@ -14,8 +14,6 @@ module ContextDev
 
     attr_reader ai: ContextDev::Resources::AI
 
-    attr_reader style: ContextDev::Resources::Style
-
     attr_reader brand: ContextDev::Resources::Brand
 
     attr_reader industry: ContextDev::Resources::Industry

data/sig/context_dev/models/{style_extract_styleguide_params.rbs → web_extract_fonts_params.rbs}

@@ -1,10 +1,10 @@
 module ContextDev
   module Models
-    type style_extract_styleguide_params =
+    type web_extract_fonts_params =
       { direct_url: String, domain: String, timeout_ms: Integer }
       & ContextDev::Internal::Type::request_parameters
 
-    class StyleExtractStyleguideParams < ContextDev::Internal::Type::BaseModel
+    class WebExtractFontsParams < ContextDev::Internal::Type::BaseModel
       extend ContextDev::Internal::Type::RequestParameters::Converter
       include ContextDev::Internal::Type::RequestParameters