context.dev 1.29.0 → 1.30.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 5d3680b28e1ece1f34117ae03f9e1338ab3ac87a3ba4abcafa312c4cb5d611fe
-  data.tar.gz: 317e0671ee43e0b05884ae45993732e3870279131ea7943523f3c9d93e490cc6
+  metadata.gz: 074be977167979846299e7e3c6eb549b3ae6ac29e1c928aec4d692111250c305
+  data.tar.gz: c6837a38aea2ae93f3faf15d835e845d5333332fc188bf92ccfa999134d33cfc
 SHA512:
-  metadata.gz: 68d2975c379e21f1916469f944acea547cd3d8cee372ef4cc911e533caa4f6ca2e661a1f0b1b42ce33b52a5a447fb0ac7caf16d51365f8c1ceef0dfa488da4b8
-  data.tar.gz: 7b34767be78b1b12b0d01c3ac4a251b4b0b59c35d53acbf029b41ca4df245469712a079e27001fba55c3ac31c39133be55e3caf4635abdeaf626c8896b9dfc4b
+  metadata.gz: 867c490615c21f264743296a01d0038280546f75eac1c7bf8355c6c32bf6fb2cf24312b66336d02298bb284db9015dd98cb9e3d1f11a588c4fd72989a8123549
+  data.tar.gz: 228294367be4138af290bfe6d0775f9266c019b4c739121f6464ad19a8998ffcb34b3b6dc377edcac4acb7a17368f211066c5499f49bd6b416c9399f15963906

data/CHANGELOG.md

@@ -1,5 +1,13 @@
 # Changelog
 
+## 1.30.0 (2026-06-07)
+
+Full Changelog: [v1.29.0...v1.30.0](https://github.com/context-dot-dev/context-ruby-sdk/compare/v1.29.0...v1.30.0)
+
+### Features
+
+* **api:** api update ([c897bbf](https://github.com/context-dot-dev/context-ruby-sdk/commit/c897bbf03235960c8036bd808c5239b935d94bf3))
+
 ## 1.29.0 (2026-06-07)
 
 Full Changelog: [v1.28.0...v1.29.0](https://github.com/context-dot-dev/context-ruby-sdk/compare/v1.28.0...v1.29.0)

data/README.md

@@ -26,7 +26,7 @@ To use this gem, install via Bundler by adding the following to your application
 <!-- x-release-please-start-version -->
 
 ```ruby
-gem "context.dev", "~> 1.29.0"
+gem "context.dev", "~> 1.30.0"
 ```
 
 <!-- x-release-please-end -->

data/lib/context_dev/models/web_web_crawl_md_params.rb

@@ -13,6 +13,14 @@ module ContextDev
       #   @return [String]
       required :url, String
 
+      # @!attribute exclude_selectors
+      #   CSS selectors to remove before each crawled page is converted to Markdown.
+      #   Applied after includeSelectors. Exclusion takes precedence: an element matching
+      #   both is removed. Examples: "nav", "footer", ".ad-banner", "[aria-hidden=true]".
+      #
+      #   @return [Array<String>, nil]
+      optional :exclude_selectors, ContextDev::Internal::Type::ArrayOf[String], api_name: :excludeSelectors
+
       # @!attribute follow_subdomains
       #   When true, follow links on subdomains of the starting URL's domain (e.g.
       #   docs.example.com when starting from example.com). www and apex are always
@@ -40,6 +48,15 @@ module ContextDev
       #   @return [Boolean, nil]
       optional :include_links, ContextDev::Internal::Type::Boolean, api_name: :includeLinks
 
+      # @!attribute include_selectors
+      #   CSS selectors. When provided, only matching HTML subtrees (and their
+      #   descendants) are kept before each crawled page is converted to Markdown. When
+      #   omitted, the entire document is kept. Examples: "article.main", "#content",
+      #   "[role=main]".
+      #
+      #   @return [Array<String>, nil]
+      optional :include_selectors, ContextDev::Internal::Type::ArrayOf[String], api_name: :includeSelectors
+
       # @!attribute max_age_ms
       #   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
@@ -110,12 +127,14 @@ module ContextDev
       #   @return [Integer, nil]
       optional :wait_for_ms, Integer, api_name: :waitForMs
 
-      # @!method initialize(url:, follow_subdomains: nil, include_frames: nil, include_images: nil, include_links: nil, max_age_ms: nil, max_depth: nil, max_pages: nil, pdf: nil, shorten_base64_images: nil, stop_after_ms: nil, timeout_ms: nil, url_regex: nil, use_main_content_only: nil, wait_for_ms: nil, request_options: {})
+      # @!method initialize(url:, exclude_selectors: nil, follow_subdomains: nil, include_frames: nil, include_images: nil, include_links: nil, include_selectors: nil, max_age_ms: nil, max_depth: nil, max_pages: nil, pdf: nil, shorten_base64_images: nil, stop_after_ms: nil, timeout_ms: nil, url_regex: nil, use_main_content_only: nil, wait_for_ms: nil, request_options: {})
       #   Some parameter documentations has been truncated, see
       #   {ContextDev::Models::WebWebCrawlMdParams} for more details.
       #
       #   @param url [String] The starting URL for the crawl (must include http:// or https:// protocol)
       #
+      #   @param exclude_selectors [Array<String>] CSS selectors to remove before each crawled page is converted to Markdown. Appli
+      #
       #   @param follow_subdomains [Boolean] When true, follow links on subdomains of the starting URL's domain (e.g. docs.ex
       #
       #   @param include_frames [Boolean] When true, the contents of iframes are rendered to Markdown for each crawled pag
@@ -124,6 +143,8 @@ module ContextDev
       #
       #   @param include_links [Boolean] Preserve hyperlinks in the Markdown output
       #
+      #   @param include_selectors [Array<String>] CSS selectors. When provided, only matching HTML subtrees (and their descendants
+      #
       #   @param max_age_ms [Integer] Return a cached result if a prior scrape for the same parameters exists and is y
       #
       #   @param max_depth [Integer] Maximum link depth from the starting URL (0 = only the starting page)

data/lib/context_dev/models/web_web_scrape_html_params.rb

@@ -13,6 +13,14 @@ module ContextDev
       #   @return [String]
       required :url, String
 
+      # @!attribute exclude_selectors
+      #   CSS selectors to remove from the result. Applied after includeSelectors.
+      #   Exclusion takes precedence: an element matching both is removed. Examples:
+      #   "nav", "footer", ".ad-banner", "[aria-hidden=true]".
+      #
+      #   @return [Array<String>, nil]
+      optional :exclude_selectors, ContextDev::Internal::Type::ArrayOf[String]
+
       # @!attribute headers
       #   Optional outbound HTTP headers forwarded only to the target URL, sent as
       #   deep-object query params such as headers[X-Custom]=value. When provided, caching
@@ -27,6 +35,14 @@ module ContextDev
       #   @return [Boolean, nil]
       optional :include_frames, ContextDev::Internal::Type::Boolean
 
+      # @!attribute include_selectors
+      #   CSS selectors. When provided, only matching subtrees (and their descendants) are
+      #   kept and everything else is dropped. When omitted, the entire document is kept.
+      #   Examples: "article.main", "#content", "[role=main]".
+      #
+      #   @return [Array<String>, nil]
+      optional :include_selectors, ContextDev::Internal::Type::ArrayOf[String]
+
       # @!attribute max_age_ms
       #   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
@@ -57,16 +73,20 @@ module ContextDev
       #   @return [Integer, nil]
       optional :wait_for_ms, Integer
 
-      # @!method initialize(url:, headers: nil, include_frames: nil, max_age_ms: nil, pdf: nil, timeout_ms: nil, wait_for_ms: nil, request_options: {})
+      # @!method initialize(url:, exclude_selectors: nil, headers: nil, include_frames: nil, include_selectors: nil, max_age_ms: nil, pdf: nil, timeout_ms: nil, wait_for_ms: nil, request_options: {})
       #   Some parameter documentations has been truncated, see
       #   {ContextDev::Models::WebWebScrapeHTMLParams} for more details.
       #
       #   @param url [String] Full URL to scrape (must include http:// or https:// protocol)
       #
+      #   @param exclude_selectors [Array<String>] CSS selectors to remove from the result. Applied after includeSelectors. Exclusi
+      #
       #   @param headers [Hash{Symbol=>String}] Optional outbound HTTP headers forwarded only to the target URL, sent as deep-ob
       #
       #   @param include_frames [Boolean] When true, iframes are rendered inline into the returned HTML.
       #
+      #   @param include_selectors [Array<String>] CSS selectors. When provided, only matching subtrees (and their descendants) are
+      #
       #   @param max_age_ms [Integer] Return a cached result if a prior scrape for the same parameters exists and is y
       #
       #   @param pdf [ContextDev::Models::WebWebScrapeHTMLParams::Pdf] PDF parsing controls. Use start/end to limit text extraction and OCR to an inclu

data/lib/context_dev/models/web_web_scrape_md_params.rb

@@ -14,6 +14,14 @@ module ContextDev
       #   @return [String]
       required :url, String
 
+      # @!attribute exclude_selectors
+      #   CSS selectors to remove before conversion to Markdown. Applied after
+      #   includeSelectors. Exclusion takes precedence: an element matching both is
+      #   removed. Examples: "nav", "footer", ".ad-banner", "[aria-hidden=true]".
+      #
+      #   @return [Array<String>, nil]
+      optional :exclude_selectors, ContextDev::Internal::Type::ArrayOf[String]
+
       # @!attribute headers
       #   Optional outbound HTTP headers forwarded only to the target URL, sent as
       #   deep-object query params such as headers[X-Custom]=value. When provided, caching
@@ -40,6 +48,14 @@ module ContextDev
       #   @return [Boolean, nil]
       optional :include_links, ContextDev::Internal::Type::Boolean
 
+      # @!attribute include_selectors
+      #   CSS selectors. When provided, only matching HTML subtrees (and their
+      #   descendants) are kept before conversion to Markdown. When omitted, the entire
+      #   document is kept. Examples: "article.main", "#content", "[role=main]".
+      #
+      #   @return [Array<String>, nil]
+      optional :include_selectors, ContextDev::Internal::Type::ArrayOf[String]
+
       # @!attribute max_age_ms
       #   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
@@ -83,12 +99,14 @@ module ContextDev
       #   @return [Integer, nil]
       optional :wait_for_ms, Integer
 
-      # @!method initialize(url:, headers: nil, include_frames: nil, include_images: nil, include_links: nil, max_age_ms: nil, pdf: nil, shorten_base64_images: nil, timeout_ms: nil, use_main_content_only: nil, wait_for_ms: nil, request_options: {})
+      # @!method initialize(url:, exclude_selectors: nil, headers: nil, include_frames: nil, include_images: nil, include_links: nil, include_selectors: nil, max_age_ms: nil, pdf: nil, shorten_base64_images: nil, timeout_ms: nil, use_main_content_only: nil, wait_for_ms: nil, request_options: {})
       #   Some parameter documentations has been truncated, see
       #   {ContextDev::Models::WebWebScrapeMdParams} for more details.
       #
       #   @param url [String] Full URL to scrape into LLM usable Markdown (must include http:// or https:// pr
       #
+      #   @param exclude_selectors [Array<String>] CSS selectors to remove before conversion to Markdown. Applied after includeSele
+      #
       #   @param headers [Hash{Symbol=>String}] Optional outbound HTTP headers forwarded only to the target URL, sent as deep-ob
       #
       #   @param include_frames [Boolean] When true, the contents of iframes are rendered to Markdown.
@@ -97,6 +115,8 @@ module ContextDev
       #
       #   @param include_links [Boolean] Preserve hyperlinks in Markdown output
       #
+      #   @param include_selectors [Array<String>] CSS selectors. When provided, only matching HTML subtrees (and their descendants
+      #
       #   @param max_age_ms [Integer] Return a cached result if a prior scrape for the same parameters exists and is y
       #
       #   @param pdf [ContextDev::Models::WebWebScrapeMdParams::Pdf] PDF parsing controls. Use start/end to limit text extraction and OCR to an inclu

data/lib/context_dev/resources/web.rb

@@ -246,10 +246,12 @@ module ContextDev
       # Performs a crawl starting from a given URL, extracts page content as Markdown,
       # and returns results for all crawled pages.
       #
-      # @overload web_crawl_md(url:, follow_subdomains: nil, include_frames: nil, include_images: nil, include_links: nil, max_age_ms: nil, max_depth: nil, max_pages: nil, pdf: nil, shorten_base64_images: nil, stop_after_ms: nil, timeout_ms: nil, url_regex: nil, use_main_content_only: nil, wait_for_ms: nil, request_options: {})
+      # @overload web_crawl_md(url:, exclude_selectors: nil, follow_subdomains: nil, include_frames: nil, include_images: nil, include_links: nil, include_selectors: nil, max_age_ms: nil, max_depth: nil, max_pages: nil, pdf: nil, shorten_base64_images: nil, stop_after_ms: nil, timeout_ms: nil, url_regex: nil, use_main_content_only: nil, wait_for_ms: nil, request_options: {})
       #
       # @param url [String] The starting URL for the crawl (must include http:// or https:// protocol)
       #
+      # @param exclude_selectors [Array<String>] CSS selectors to remove before each crawled page is converted to Markdown. Appli
+      #
       # @param follow_subdomains [Boolean] When true, follow links on subdomains of the starting URL's domain (e.g. docs.ex
       #
       # @param include_frames [Boolean] When true, the contents of iframes are rendered to Markdown for each crawled pag
@@ -258,6 +260,8 @@ module ContextDev
       #
       # @param include_links [Boolean] Preserve hyperlinks in the Markdown output
       #
+      # @param include_selectors [Array<String>] CSS selectors. When provided, only matching HTML subtrees (and their descendants
+      #
       # @param max_age_ms [Integer] Return a cached result if a prior scrape for the same parameters exists and is y
       #
       # @param max_depth [Integer] Maximum link depth from the starting URL (0 = only the starting page)
@@ -299,14 +303,18 @@ module ContextDev
       #
       # Scrapes the given URL and returns the raw HTML content of the page.
       #
-      # @overload web_scrape_html(url:, headers: nil, include_frames: nil, max_age_ms: nil, pdf: nil, timeout_ms: nil, wait_for_ms: nil, request_options: {})
+      # @overload web_scrape_html(url:, exclude_selectors: nil, headers: nil, include_frames: nil, include_selectors: nil, max_age_ms: nil, pdf: nil, timeout_ms: nil, wait_for_ms: nil, request_options: {})
       #
       # @param url [String] Full URL to scrape (must include http:// or https:// protocol)
       #
+      # @param exclude_selectors [Array<String>] CSS selectors to remove from the result. Applied after includeSelectors. Exclusi
+      #
       # @param headers [Hash{Symbol=>String}] Optional outbound HTTP headers forwarded only to the target URL, sent as deep-ob
       #
       # @param include_frames [Boolean] When true, iframes are rendered inline into the returned HTML.
       #
+      # @param include_selectors [Array<String>] CSS selectors. When provided, only matching subtrees (and their descendants) are
+      #
       # @param max_age_ms [Integer] Return a cached result if a prior scrape for the same parameters exists and is y
       #
       # @param pdf [ContextDev::Models::WebWebScrapeHTMLParams::Pdf] PDF parsing controls. Use start/end to limit text extraction and OCR to an inclu
@@ -327,7 +335,9 @@ module ContextDev
           method: :get,
           path: "web/scrape/html",
           query: query.transform_keys(
+            exclude_selectors: "excludeSelectors",
             include_frames: "includeFrames",
+            include_selectors: "includeSelectors",
             max_age_ms: "maxAgeMs",
             timeout_ms: "timeoutMS",
             wait_for_ms: "waitForMs"
@@ -385,10 +395,12 @@ module ContextDev
       #
       # Scrapes the given URL into LLM usable Markdown.
       #
-      # @overload web_scrape_md(url:, headers: nil, include_frames: nil, include_images: nil, include_links: nil, max_age_ms: nil, pdf: nil, shorten_base64_images: nil, timeout_ms: nil, use_main_content_only: nil, wait_for_ms: nil, request_options: {})
+      # @overload web_scrape_md(url:, exclude_selectors: nil, headers: nil, include_frames: nil, include_images: nil, include_links: nil, include_selectors: nil, max_age_ms: nil, pdf: nil, shorten_base64_images: nil, timeout_ms: nil, use_main_content_only: nil, wait_for_ms: nil, request_options: {})
       #
       # @param url [String] Full URL to scrape into LLM usable Markdown (must include http:// or https:// pr
       #
+      # @param exclude_selectors [Array<String>] CSS selectors to remove before conversion to Markdown. Applied after includeSele
+      #
       # @param headers [Hash{Symbol=>String}] Optional outbound HTTP headers forwarded only to the target URL, sent as deep-ob
       #
       # @param include_frames [Boolean] When true, the contents of iframes are rendered to Markdown.
@@ -397,6 +409,8 @@ module ContextDev
       #
       # @param include_links [Boolean] Preserve hyperlinks in Markdown output
       #
+      # @param include_selectors [Array<String>] CSS selectors. When provided, only matching HTML subtrees (and their descendants
+      #
       # @param max_age_ms [Integer] Return a cached result if a prior scrape for the same parameters exists and is y
       #
       # @param pdf [ContextDev::Models::WebWebScrapeMdParams::Pdf] PDF parsing controls. Use start/end to limit text extraction and OCR to an inclu
@@ -421,9 +435,11 @@ module ContextDev
           method: :get,
           path: "web/scrape/markdown",
           query: query.transform_keys(
+            exclude_selectors: "excludeSelectors",
             include_frames: "includeFrames",
             include_images: "includeImages",
             include_links: "includeLinks",
+            include_selectors: "includeSelectors",
             max_age_ms: "maxAgeMs",
             shorten_base64_images: "shortenBase64Images",
             timeout_ms: "timeoutMS",

data/lib/context_dev/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module ContextDev
-  VERSION = "1.29.0"
+  VERSION = "1.30.0"
 end

data/rbi/context_dev/models/web_web_crawl_md_params.rbi

@@ -15,6 +15,15 @@ module ContextDev
       sig { returns(String) }
       attr_accessor :url
 
+      # CSS selectors to remove before each crawled page is converted to Markdown.
+      # Applied after includeSelectors. Exclusion takes precedence: an element matching
+      # both is removed. Examples: "nav", "footer", ".ad-banner", "[aria-hidden=true]".
+      sig { returns(T.nilable(T::Array[String])) }
+      attr_reader :exclude_selectors
+
+      sig { params(exclude_selectors: T::Array[String]).void }
+      attr_writer :exclude_selectors
+
       # When true, follow links on subdomains of the starting URL's domain (e.g.
       # docs.example.com when starting from example.com). www and apex are always
       # treated as equivalent.
@@ -46,6 +55,16 @@ module ContextDev
       sig { params(include_links: T::Boolean).void }
       attr_writer :include_links
 
+      # CSS selectors. When provided, only matching HTML subtrees (and their
+      # descendants) are kept before each crawled page is converted to Markdown. When
+      # omitted, the entire document is kept. Examples: "article.main", "#content",
+      # "[role=main]".
+      sig { returns(T.nilable(T::Array[String])) }
+      attr_reader :include_selectors
+
+      sig { params(include_selectors: T::Array[String]).void }
+      attr_writer :include_selectors
+
       # 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.
@@ -129,10 +148,12 @@ module ContextDev
       sig do
         params(
           url: String,
+          exclude_selectors: T::Array[String],
           follow_subdomains: T::Boolean,
           include_frames: T::Boolean,
           include_images: T::Boolean,
           include_links: T::Boolean,
+          include_selectors: T::Array[String],
           max_age_ms: Integer,
           max_depth: Integer,
           max_pages: Integer,
@@ -149,6 +170,10 @@ module ContextDev
       def self.new(
         # The starting URL for the crawl (must include http:// or https:// protocol)
         url:,
+        # CSS selectors to remove before each crawled page is converted to Markdown.
+        # Applied after includeSelectors. Exclusion takes precedence: an element matching
+        # both is removed. Examples: "nav", "footer", ".ad-banner", "[aria-hidden=true]".
+        exclude_selectors: nil,
         # When true, follow links on subdomains of the starting URL's domain (e.g.
         # docs.example.com when starting from example.com). www and apex are always
         # treated as equivalent.
@@ -160,6 +185,11 @@ module ContextDev
         include_images: nil,
         # Preserve hyperlinks in the Markdown output
         include_links: nil,
+        # CSS selectors. When provided, only matching HTML subtrees (and their
+        # descendants) are kept before each crawled page is converted to Markdown. When
+        # omitted, the entire document is kept. Examples: "article.main", "#content",
+        # "[role=main]".
+        include_selectors: 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.
@@ -198,10 +228,12 @@ module ContextDev
         override.returns(
           {
             url: String,
+            exclude_selectors: T::Array[String],
             follow_subdomains: T::Boolean,
             include_frames: T::Boolean,
             include_images: T::Boolean,
             include_links: T::Boolean,
+            include_selectors: T::Array[String],
             max_age_ms: Integer,
             max_depth: Integer,
             max_pages: Integer,

data/rbi/context_dev/models/web_web_scrape_html_params.rbi

@@ -18,6 +18,15 @@ module ContextDev
       sig { returns(String) }
       attr_accessor :url
 
+      # CSS selectors to remove from the result. Applied after includeSelectors.
+      # Exclusion takes precedence: an element matching both is removed. Examples:
+      # "nav", "footer", ".ad-banner", "[aria-hidden=true]".
+      sig { returns(T.nilable(T::Array[String])) }
+      attr_reader :exclude_selectors
+
+      sig { params(exclude_selectors: T::Array[String]).void }
+      attr_writer :exclude_selectors
+
       # Optional outbound HTTP headers forwarded only to the target URL, sent as
       # deep-object query params such as headers[X-Custom]=value. When provided, caching
       # is bypassed: the result is neither read from nor written to cache.
@@ -34,6 +43,15 @@ module ContextDev
       sig { params(include_frames: T::Boolean).void }
       attr_writer :include_frames
 
+      # CSS selectors. When provided, only matching subtrees (and their descendants) are
+      # kept and everything else is dropped. When omitted, the entire document is kept.
+      # Examples: "article.main", "#content", "[role=main]".
+      sig { returns(T.nilable(T::Array[String])) }
+      attr_reader :include_selectors
+
+      sig { params(include_selectors: T::Array[String]).void }
+      attr_writer :include_selectors
+
       # 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.
@@ -71,8 +89,10 @@ module ContextDev
       sig do
         params(
           url: String,
+          exclude_selectors: T::Array[String],
           headers: T::Hash[Symbol, String],
           include_frames: T::Boolean,
+          include_selectors: T::Array[String],
           max_age_ms: Integer,
           pdf: ContextDev::WebWebScrapeHTMLParams::Pdf::OrHash,
           timeout_ms: Integer,
@@ -83,12 +103,20 @@ module ContextDev
       def self.new(
         # Full URL to scrape (must include http:// or https:// protocol)
         url:,
+        # CSS selectors to remove from the result. Applied after includeSelectors.
+        # Exclusion takes precedence: an element matching both is removed. Examples:
+        # "nav", "footer", ".ad-banner", "[aria-hidden=true]".
+        exclude_selectors: nil,
         # Optional outbound HTTP headers forwarded only to the target URL, sent as
         # deep-object query params such as headers[X-Custom]=value. When provided, caching
         # is bypassed: the result is neither read from nor written to cache.
         headers: nil,
         # When true, iframes are rendered inline into the returned HTML.
         include_frames: nil,
+        # CSS selectors. When provided, only matching subtrees (and their descendants) are
+        # kept and everything else is dropped. When omitted, the entire document is kept.
+        # Examples: "article.main", "#content", "[role=main]".
+        include_selectors: 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.
@@ -111,8 +139,10 @@ module ContextDev
         override.returns(
           {
             url: String,
+            exclude_selectors: T::Array[String],
             headers: T::Hash[Symbol, String],
             include_frames: T::Boolean,
+            include_selectors: T::Array[String],
             max_age_ms: Integer,
             pdf: ContextDev::WebWebScrapeHTMLParams::Pdf,
             timeout_ms: Integer,

data/rbi/context_dev/models/web_web_scrape_md_params.rbi

@@ -16,6 +16,15 @@ module ContextDev
       sig { returns(String) }
       attr_accessor :url
 
+      # CSS selectors to remove before conversion to Markdown. Applied after
+      # includeSelectors. Exclusion takes precedence: an element matching both is
+      # removed. Examples: "nav", "footer", ".ad-banner", "[aria-hidden=true]".
+      sig { returns(T.nilable(T::Array[String])) }
+      attr_reader :exclude_selectors
+
+      sig { params(exclude_selectors: T::Array[String]).void }
+      attr_writer :exclude_selectors
+
       # Optional outbound HTTP headers forwarded only to the target URL, sent as
       # deep-object query params such as headers[X-Custom]=value. When provided, caching
       # is bypassed: the result is neither read from nor written to cache.
@@ -46,6 +55,15 @@ module ContextDev
       sig { params(include_links: T::Boolean).void }
       attr_writer :include_links
 
+      # CSS selectors. When provided, only matching HTML subtrees (and their
+      # descendants) are kept before conversion to Markdown. When omitted, the entire
+      # document is kept. Examples: "article.main", "#content", "[role=main]".
+      sig { returns(T.nilable(T::Array[String])) }
+      attr_reader :include_selectors
+
+      sig { params(include_selectors: T::Array[String]).void }
+      attr_writer :include_selectors
+
       # 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.
@@ -98,10 +116,12 @@ module ContextDev
       sig do
         params(
           url: String,
+          exclude_selectors: T::Array[String],
           headers: T::Hash[Symbol, String],
           include_frames: T::Boolean,
           include_images: T::Boolean,
           include_links: T::Boolean,
+          include_selectors: T::Array[String],
           max_age_ms: Integer,
           pdf: ContextDev::WebWebScrapeMdParams::Pdf::OrHash,
           shorten_base64_images: T::Boolean,
@@ -115,6 +135,10 @@ module ContextDev
         # Full URL to scrape into LLM usable Markdown (must include http:// or https://
         # protocol)
         url:,
+        # CSS selectors to remove before conversion to Markdown. Applied after
+        # includeSelectors. Exclusion takes precedence: an element matching both is
+        # removed. Examples: "nav", "footer", ".ad-banner", "[aria-hidden=true]".
+        exclude_selectors: nil,
         # Optional outbound HTTP headers forwarded only to the target URL, sent as
         # deep-object query params such as headers[X-Custom]=value. When provided, caching
         # is bypassed: the result is neither read from nor written to cache.
@@ -125,6 +149,10 @@ module ContextDev
         include_images: nil,
         # Preserve hyperlinks in Markdown output
         include_links: nil,
+        # CSS selectors. When provided, only matching HTML subtrees (and their
+        # descendants) are kept before conversion to Markdown. When omitted, the entire
+        # document is kept. Examples: "article.main", "#content", "[role=main]".
+        include_selectors: 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.
@@ -152,10 +180,12 @@ module ContextDev
         override.returns(
           {
             url: String,
+            exclude_selectors: T::Array[String],
             headers: T::Hash[Symbol, String],
             include_frames: T::Boolean,
             include_images: T::Boolean,
             include_links: T::Boolean,
+            include_selectors: T::Array[String],
             max_age_ms: Integer,
             pdf: ContextDev::WebWebScrapeMdParams::Pdf,
             shorten_base64_images: T::Boolean,

data/rbi/context_dev/resources/web.rbi

@@ -251,10 +251,12 @@ module ContextDev
       sig do
         params(
           url: String,
+          exclude_selectors: T::Array[String],
           follow_subdomains: T::Boolean,
           include_frames: T::Boolean,
           include_images: T::Boolean,
           include_links: T::Boolean,
+          include_selectors: T::Array[String],
           max_age_ms: Integer,
           max_depth: Integer,
           max_pages: Integer,
@@ -271,6 +273,10 @@ module ContextDev
       def web_crawl_md(
         # The starting URL for the crawl (must include http:// or https:// protocol)
         url:,
+        # CSS selectors to remove before each crawled page is converted to Markdown.
+        # Applied after includeSelectors. Exclusion takes precedence: an element matching
+        # both is removed. Examples: "nav", "footer", ".ad-banner", "[aria-hidden=true]".
+        exclude_selectors: nil,
         # When true, follow links on subdomains of the starting URL's domain (e.g.
         # docs.example.com when starting from example.com). www and apex are always
         # treated as equivalent.
@@ -282,6 +288,11 @@ module ContextDev
         include_images: nil,
         # Preserve hyperlinks in the Markdown output
         include_links: nil,
+        # CSS selectors. When provided, only matching HTML subtrees (and their
+        # descendants) are kept before each crawled page is converted to Markdown. When
+        # omitted, the entire document is kept. Examples: "article.main", "#content",
+        # "[role=main]".
+        include_selectors: 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.
@@ -320,8 +331,10 @@ module ContextDev
       sig do
         params(
           url: String,
+          exclude_selectors: T::Array[String],
           headers: T::Hash[Symbol, String],
           include_frames: T::Boolean,
+          include_selectors: T::Array[String],
           max_age_ms: Integer,
           pdf: ContextDev::WebWebScrapeHTMLParams::Pdf::OrHash,
           timeout_ms: Integer,
@@ -332,12 +345,20 @@ module ContextDev
       def web_scrape_html(
         # Full URL to scrape (must include http:// or https:// protocol)
         url:,
+        # CSS selectors to remove from the result. Applied after includeSelectors.
+        # Exclusion takes precedence: an element matching both is removed. Examples:
+        # "nav", "footer", ".ad-banner", "[aria-hidden=true]".
+        exclude_selectors: nil,
         # Optional outbound HTTP headers forwarded only to the target URL, sent as
         # deep-object query params such as headers[X-Custom]=value. When provided, caching
         # is bypassed: the result is neither read from nor written to cache.
         headers: nil,
         # When true, iframes are rendered inline into the returned HTML.
         include_frames: nil,
+        # CSS selectors. When provided, only matching subtrees (and their descendants) are
+        # kept and everything else is dropped. When omitted, the entire document is kept.
+        # Examples: "article.main", "#content", "[role=main]".
+        include_selectors: 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.
@@ -399,10 +420,12 @@ module ContextDev
       sig do
         params(
           url: String,
+          exclude_selectors: T::Array[String],
           headers: T::Hash[Symbol, String],
           include_frames: T::Boolean,
           include_images: T::Boolean,
           include_links: T::Boolean,
+          include_selectors: T::Array[String],
           max_age_ms: Integer,
           pdf: ContextDev::WebWebScrapeMdParams::Pdf::OrHash,
           shorten_base64_images: T::Boolean,
@@ -416,6 +439,10 @@ module ContextDev
         # Full URL to scrape into LLM usable Markdown (must include http:// or https://
         # protocol)
         url:,
+        # CSS selectors to remove before conversion to Markdown. Applied after
+        # includeSelectors. Exclusion takes precedence: an element matching both is
+        # removed. Examples: "nav", "footer", ".ad-banner", "[aria-hidden=true]".
+        exclude_selectors: nil,
         # Optional outbound HTTP headers forwarded only to the target URL, sent as
         # deep-object query params such as headers[X-Custom]=value. When provided, caching
         # is bypassed: the result is neither read from nor written to cache.
@@ -426,6 +453,10 @@ module ContextDev
         include_images: nil,
         # Preserve hyperlinks in Markdown output
         include_links: nil,
+        # CSS selectors. When provided, only matching HTML subtrees (and their
+        # descendants) are kept before conversion to Markdown. When omitted, the entire
+        # document is kept. Examples: "article.main", "#content", "[role=main]".
+        include_selectors: 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.

data/sig/context_dev/models/web_web_crawl_md_params.rbs

@@ -3,10 +3,12 @@ module ContextDev
     type web_web_crawl_md_params =
       {
         url: String,
+        exclude_selectors: ::Array[String],
         follow_subdomains: bool,
         include_frames: bool,
         include_images: bool,
         include_links: bool,
+        include_selectors: ::Array[String],
         max_age_ms: Integer,
         max_depth: Integer,
         max_pages: Integer,
@@ -26,6 +28,10 @@ module ContextDev
 
       attr_accessor url: String
 
+      attr_reader exclude_selectors: ::Array[String]?
+
+      def exclude_selectors=: (::Array[String]) -> ::Array[String]
+
       attr_reader follow_subdomains: bool?
 
       def follow_subdomains=: (bool) -> bool
@@ -42,6 +48,10 @@ module ContextDev
 
       def include_links=: (bool) -> bool
 
+      attr_reader include_selectors: ::Array[String]?
+
+      def include_selectors=: (::Array[String]) -> ::Array[String]
+
       attr_reader max_age_ms: Integer?
 
       def max_age_ms=: (Integer) -> Integer
@@ -86,10 +96,12 @@ module ContextDev
 
       def initialize: (
         url: String,
+        ?exclude_selectors: ::Array[String],
         ?follow_subdomains: bool,
         ?include_frames: bool,
         ?include_images: bool,
         ?include_links: bool,
+        ?include_selectors: ::Array[String],
         ?max_age_ms: Integer,
         ?max_depth: Integer,
         ?max_pages: Integer,
@@ -105,10 +117,12 @@ module ContextDev
 
       def to_hash: -> {
         url: String,
+        exclude_selectors: ::Array[String],
         follow_subdomains: bool,
         include_frames: bool,
         include_images: bool,
         include_links: bool,
+        include_selectors: ::Array[String],
         max_age_ms: Integer,
         max_depth: Integer,
         max_pages: Integer,

data/sig/context_dev/models/web_web_scrape_html_params.rbs

@@ -3,8 +3,10 @@ module ContextDev
     type web_web_scrape_html_params =
       {
         url: String,
+        exclude_selectors: ::Array[String],
         headers: ::Hash[Symbol, String],
         include_frames: bool,
+        include_selectors: ::Array[String],
         max_age_ms: Integer,
         pdf: ContextDev::WebWebScrapeHTMLParams::Pdf,
         timeout_ms: Integer,
@@ -18,6 +20,10 @@ module ContextDev
 
       attr_accessor url: String
 
+      attr_reader exclude_selectors: ::Array[String]?
+
+      def exclude_selectors=: (::Array[String]) -> ::Array[String]
+
       attr_reader headers: ::Hash[Symbol, String]?
 
       def headers=: (::Hash[Symbol, String]) -> ::Hash[Symbol, String]
@@ -26,6 +32,10 @@ module ContextDev
 
       def include_frames=: (bool) -> bool
 
+      attr_reader include_selectors: ::Array[String]?
+
+      def include_selectors=: (::Array[String]) -> ::Array[String]
+
       attr_reader max_age_ms: Integer?
 
       def max_age_ms=: (Integer) -> Integer
@@ -46,8 +56,10 @@ module ContextDev
 
       def initialize: (
         url: String,
+        ?exclude_selectors: ::Array[String],
         ?headers: ::Hash[Symbol, String],
         ?include_frames: bool,
+        ?include_selectors: ::Array[String],
         ?max_age_ms: Integer,
         ?pdf: ContextDev::WebWebScrapeHTMLParams::Pdf,
         ?timeout_ms: Integer,
@@ -57,8 +69,10 @@ module ContextDev
 
       def to_hash: -> {
         url: String,
+        exclude_selectors: ::Array[String],
         headers: ::Hash[Symbol, String],
         include_frames: bool,
+        include_selectors: ::Array[String],
         max_age_ms: Integer,
         pdf: ContextDev::WebWebScrapeHTMLParams::Pdf,
         timeout_ms: Integer,

data/sig/context_dev/models/web_web_scrape_md_params.rbs

@@ -3,10 +3,12 @@ module ContextDev
     type web_web_scrape_md_params =
       {
         url: String,
+        exclude_selectors: ::Array[String],
         headers: ::Hash[Symbol, String],
         include_frames: bool,
         include_images: bool,
         include_links: bool,
+        include_selectors: ::Array[String],
         max_age_ms: Integer,
         pdf: ContextDev::WebWebScrapeMdParams::Pdf,
         :shorten_base64_images => bool,
@@ -22,6 +24,10 @@ module ContextDev
 
       attr_accessor url: String
 
+      attr_reader exclude_selectors: ::Array[String]?
+
+      def exclude_selectors=: (::Array[String]) -> ::Array[String]
+
       attr_reader headers: ::Hash[Symbol, String]?
 
       def headers=: (::Hash[Symbol, String]) -> ::Hash[Symbol, String]
@@ -38,6 +44,10 @@ module ContextDev
 
       def include_links=: (bool) -> bool
 
+      attr_reader include_selectors: ::Array[String]?
+
+      def include_selectors=: (::Array[String]) -> ::Array[String]
+
       attr_reader max_age_ms: Integer?
 
       def max_age_ms=: (Integer) -> Integer
@@ -66,10 +76,12 @@ module ContextDev
 
       def initialize: (
         url: String,
+        ?exclude_selectors: ::Array[String],
         ?headers: ::Hash[Symbol, String],
         ?include_frames: bool,
         ?include_images: bool,
         ?include_links: bool,
+        ?include_selectors: ::Array[String],
         ?max_age_ms: Integer,
         ?pdf: ContextDev::WebWebScrapeMdParams::Pdf,
         ?shorten_base64_images: bool,
@@ -81,10 +93,12 @@ module ContextDev
 
       def to_hash: -> {
         url: String,
+        exclude_selectors: ::Array[String],
         headers: ::Hash[Symbol, String],
         include_frames: bool,
         include_images: bool,
         include_links: bool,
+        include_selectors: ::Array[String],
         max_age_ms: Integer,
         pdf: ContextDev::WebWebScrapeMdParams::Pdf,
         :shorten_base64_images => bool,

data/sig/context_dev/resources/web.rbs

@@ -65,10 +65,12 @@ module ContextDev
 
       def web_crawl_md: (
         url: String,
+        ?exclude_selectors: ::Array[String],
         ?follow_subdomains: bool,
         ?include_frames: bool,
         ?include_images: bool,
         ?include_links: bool,
+        ?include_selectors: ::Array[String],
         ?max_age_ms: Integer,
         ?max_depth: Integer,
         ?max_pages: Integer,
@@ -84,8 +86,10 @@ module ContextDev
 
       def web_scrape_html: (
         url: String,
+        ?exclude_selectors: ::Array[String],
         ?headers: ::Hash[Symbol, String],
         ?include_frames: bool,
+        ?include_selectors: ::Array[String],
         ?max_age_ms: Integer,
         ?pdf: ContextDev::WebWebScrapeHTMLParams::Pdf,
         ?timeout_ms: Integer,
@@ -105,10 +109,12 @@ module ContextDev
 
       def web_scrape_md: (
         url: String,
+        ?exclude_selectors: ::Array[String],
         ?headers: ::Hash[Symbol, String],
         ?include_frames: bool,
         ?include_images: bool,
         ?include_links: bool,
+        ?include_selectors: ::Array[String],
         ?max_age_ms: Integer,
         ?pdf: ContextDev::WebWebScrapeMdParams::Pdf,
         ?shorten_base64_images: bool,

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: context.dev
 version: !ruby/object:Gem::Version
-  version: 1.29.0
+  version: 1.30.0
 platform: ruby
 authors:
 - Context Dev