context.dev 1.20.1 → 1.21.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 3ab6a6538a4daaf8da84619991d0ab42b69981fddc96323e3525f040b19f4e26
-  data.tar.gz: 8844bde32208534b85cf200b78984579a53eb7580e65003a024c85038699daaa
+  metadata.gz: 07c029ca2179a5702e57c1770842bbca532868ee3c9c70316c5eb7e6be31d194
+  data.tar.gz: a010b812638e6fe9907811b9dc26e9c8d95bda8b98321669b015409be72437bd
 SHA512:
-  metadata.gz: 86b2070b828fe249ccfa9e10fd50ac470473ac89b59023f88404eed9d51a66b1a2d1578425151327d8ce83ae9f78933e80875a41add0e0d4a0462697c6a8cd97
-  data.tar.gz: bd7f7c59e2a4a31a81801c6d6b80ebe0b49faa48455f86d83a228b5210ad17caaad4a861291f8e0e441288227bfbeae8cbfee7f92416f15bb0e8b9f18f515b60
+  metadata.gz: 7d0670b26708d40b8d81280ec687971211d1c751b4690fb9eafeaf193a60b74f71f82628c874263255e17c6bc632bd62b5afb13b6d30039664a2c33bc9777543
+  data.tar.gz: 7a32179cbdc288d15e7d138083cb8d26eddddc80525a5125ff40e61183a60b1fdefa766a850837abd2dc83980219ac514718367b1e34de82c4f96e50d5aca779

data/CHANGELOG.md

@@ -1,5 +1,13 @@
 # Changelog
 
+## 1.21.0 (2026-05-16)
+
+Full Changelog: [v1.20.1...v1.21.0](https://github.com/context-dot-dev/context-ruby-sdk/compare/v1.20.1...v1.21.0)
+
+### Features
+
+* **api:** manual updates ([b6c283e](https://github.com/context-dot-dev/context-ruby-sdk/commit/b6c283ee6c514421e07fb0b6d61aace62df318d1))
+
 ## 1.20.1 (2026-05-14)
 
 Full Changelog: [v1.20.0...v1.20.1](https://github.com/context-dot-dev/context-ruby-sdk/compare/v1.20.0...v1.20.1)

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", "~> 1.20.1"
+gem "context.dev", "~> 1.21.0"
 ```
 
 <!-- x-release-please-end -->

data/lib/context_dev/models/web_search_params.rb

@@ -0,0 +1,218 @@
+# frozen_string_literal: true
+
+module ContextDev
+  module Models
+    # @see ContextDev::Resources::Web#search
+    class WebSearchParams < ContextDev::Internal::Type::BaseModel
+      extend ContextDev::Internal::Type::RequestParameters::Converter
+      include ContextDev::Internal::Type::RequestParameters
+
+      # @!attribute query
+      #   Natural-language search query.
+      #
+      #   @return [String]
+      required :query, String
+
+      # @!attribute exclude_domains
+      #   Blocklist — drop results from these domains. Example: ["pinterest.com",
+      #   "reddit.com"].
+      #
+      #   @return [Array<String>, nil]
+      optional :exclude_domains, ContextDev::Internal::Type::ArrayOf[String], api_name: :excludeDomains
+
+      # @!attribute freshness
+      #   Restrict results to content published within this window.
+      #
+      #   @return [Symbol, ContextDev::Models::WebSearchParams::Freshness, nil]
+      optional :freshness, enum: -> { ContextDev::WebSearchParams::Freshness }
+
+      # @!attribute include_domains
+      #   Allowlist — only return results from these domains. Example: ["arxiv.org",
+      #   "github.com"].
+      #
+      #   @return [Array<String>, nil]
+      optional :include_domains, ContextDev::Internal::Type::ArrayOf[String], api_name: :includeDomains
+
+      # @!attribute markdown_options
+      #   Inline Markdown scraping for each result. Set `enabled: true` to activate.
+      #
+      #   @return [ContextDev::Models::WebSearchParams::MarkdownOptions, nil]
+      optional :markdown_options,
+               -> { ContextDev::WebSearchParams::MarkdownOptions },
+               api_name: :markdownOptions
+
+      # @!attribute query_fanout
+      #   Expand the query into multiple parallel variants for broader recall.
+      #
+      #   @return [Boolean, nil]
+      optional :query_fanout, ContextDev::Internal::Type::Boolean, api_name: :queryFanout
+
+      # @!attribute timeout_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).
+      #
+      #   @return [Integer, nil]
+      optional :timeout_ms, Integer, api_name: :timeoutMS
+
+      # @!method initialize(query:, exclude_domains: nil, freshness: nil, include_domains: nil, markdown_options: nil, query_fanout: nil, timeout_ms: nil, request_options: {})
+      #   Some parameter documentations has been truncated, see
+      #   {ContextDev::Models::WebSearchParams} for more details.
+      #
+      #   @param query [String] Natural-language search query.
+      #
+      #   @param exclude_domains [Array<String>] Blocklist — drop results from these domains. Example: ["pinterest.com", "reddit.
+      #
+      #   @param freshness [Symbol, ContextDev::Models::WebSearchParams::Freshness] Restrict results to content published within this window.
+      #
+      #   @param include_domains [Array<String>] Allowlist — only return results from these domains. Example: ["arxiv.org", "gith
+      #
+      #   @param markdown_options [ContextDev::Models::WebSearchParams::MarkdownOptions] Inline Markdown scraping for each result. Set `enabled: true` to activate.
+      #
+      #   @param query_fanout [Boolean] Expand the query into multiple parallel variants for broader recall.
+      #
+      #   @param timeout_ms [Integer] Optional timeout in milliseconds for the request. If the request takes longer th
+      #
+      #   @param request_options [ContextDev::RequestOptions, Hash{Symbol=>Object}]
+
+      # Restrict results to content published within this window.
+      module Freshness
+        extend ContextDev::Internal::Type::Enum
+
+        LAST_24_HOURS = :last_24_hours
+        LAST_WEEK = :last_week
+        LAST_MONTH = :last_month
+        LAST_YEAR = :last_year
+
+        # @!method self.values
+        #   @return [Array<Symbol>]
+      end
+
+      class MarkdownOptions < ContextDev::Internal::Type::BaseModel
+        # @!attribute enabled
+        #   Scrape each result to Markdown. Off by default to keep search cheap and fast.
+        #
+        #   @return [Boolean, nil]
+        optional :enabled, ContextDev::Internal::Type::Boolean
+
+        # @!attribute include_frames
+        #   Render iframe contents into the Markdown.
+        #
+        #   @return [Boolean, nil]
+        optional :include_frames, ContextDev::Internal::Type::Boolean, api_name: :includeFrames
+
+        # @!attribute include_images
+        #   Emit image references in the Markdown.
+        #
+        #   @return [Boolean, nil]
+        optional :include_images, ContextDev::Internal::Type::Boolean, api_name: :includeImages
+
+        # @!attribute include_links
+        #   Keep hyperlinks in the Markdown.
+        #
+        #   @return [Boolean, nil]
+        optional :include_links, ContextDev::Internal::Type::Boolean, api_name: :includeLinks
+
+        # @!attribute max_age_ms
+        #   Cache TTL in ms for scraped Markdown keyed by URL + options. Default 1 day, max
+        #   30 days. Set to 0 to force a fresh scrape.
+        #
+        #   @return [Integer, nil]
+        optional :max_age_ms, Integer, api_name: :maxAgeMs
+
+        # @!attribute pdf
+        #   PDF handling. Use start/end to bound text extraction and OCR to a page range.
+        #
+        #   @return [ContextDev::Models::WebSearchParams::MarkdownOptions::Pdf, nil]
+        optional :pdf, -> { ContextDev::WebSearchParams::MarkdownOptions::Pdf }
+
+        # @!attribute shorten_base64_images
+        #   Truncate inline base64 image payloads to keep responses small.
+        #
+        #   @return [Boolean, nil]
+        optional :shorten_base64_images, ContextDev::Internal::Type::Boolean, api_name: :shortenBase64Images
+
+        # @!attribute timeout_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).
+        #
+        #   @return [Integer, nil]
+        optional :timeout_ms, Integer, api_name: :timeoutMS
+
+        # @!attribute use_main_content_only
+        #   Strip nav, header, footer, and sidebar — keep only the primary article content.
+        #
+        #   @return [Boolean, nil]
+        optional :use_main_content_only, ContextDev::Internal::Type::Boolean, api_name: :useMainContentOnly
+
+        # @!attribute wait_for_ms
+        #   Extra wait after page load before rendering, in ms (0–30000). Useful for
+        #   JS-heavy pages.
+        #
+        #   @return [Integer, nil]
+        optional :wait_for_ms, Integer, api_name: :waitForMs
+
+        # @!method initialize(enabled: 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)
+        #   Some parameter documentations has been truncated, see
+        #   {ContextDev::Models::WebSearchParams::MarkdownOptions} for more details.
+        #
+        #   Inline Markdown scraping for each result. Set `enabled: true` to activate.
+        #
+        #   @param enabled [Boolean] Scrape each result to Markdown. Off by default to keep search cheap and fast.
+        #
+        #   @param include_frames [Boolean] Render iframe contents into the Markdown.
+        #
+        #   @param include_images [Boolean] Emit image references in the Markdown.
+        #
+        #   @param include_links [Boolean] Keep hyperlinks in the Markdown.
+        #
+        #   @param max_age_ms [Integer] Cache TTL in ms for scraped Markdown keyed by URL + options. Default 1 day, max
+        #
+        #   @param pdf [ContextDev::Models::WebSearchParams::MarkdownOptions::Pdf] PDF handling. Use start/end to bound text extraction and OCR to a page range.
+        #
+        #   @param shorten_base64_images [Boolean] Truncate inline base64 image payloads to keep responses small.
+        #
+        #   @param timeout_ms [Integer] Optional timeout in milliseconds for the request. If the request takes longer th
+        #
+        #   @param use_main_content_only [Boolean] Strip nav, header, footer, and sidebar — keep only the primary article content.
+        #
+        #   @param wait_for_ms [Integer] Extra wait after page load before rendering, in ms (0–30000). Useful for JS-heav
+
+        # @see ContextDev::Models::WebSearchParams::MarkdownOptions#pdf
+        class Pdf < ContextDev::Internal::Type::BaseModel
+          # @!attribute end_
+          #   Last PDF page to parse (1-based, inclusive). Defaults to the final page. Must
+          #   be >= start.
+          #
+          #   @return [Integer, nil]
+          optional :end_, Integer, api_name: :end
+
+          # @!attribute should_parse
+          #   Parse PDF URLs. When false, PDF results are skipped with WEBSITE_ACCESS_ERROR.
+          #
+          #   @return [Boolean, nil]
+          optional :should_parse, ContextDev::Internal::Type::Boolean, api_name: :shouldParse
+
+          # @!attribute start
+          #   First PDF page to parse (1-based, inclusive). Defaults to page 1.
+          #
+          #   @return [Integer, nil]
+          optional :start, Integer
+
+          # @!method initialize(end_: nil, should_parse: nil, start: nil)
+          #   Some parameter documentations has been truncated, see
+          #   {ContextDev::Models::WebSearchParams::MarkdownOptions::Pdf} for more details.
+          #
+          #   PDF handling. Use start/end to bound text extraction and OCR to a page range.
+          #
+          #   @param end_ [Integer] Last PDF page to parse (1-based, inclusive). Defaults to the final page. Must be
+          #
+          #   @param should_parse [Boolean] Parse PDF URLs. When false, PDF results are skipped with WEBSITE_ACCESS_ERROR.
+          #
+          #   @param start [Integer] First PDF page to parse (1-based, inclusive). Defaults to page 1.
+        end
+      end
+    end
+  end
+end

data/lib/context_dev/models/web_search_response.rb

@@ -0,0 +1,124 @@
+# frozen_string_literal: true
+
+module ContextDev
+  module Models
+    # @see ContextDev::Resources::Web#search
+    class WebSearchResponse < ContextDev::Internal::Type::BaseModel
+      # @!attribute query
+      #   Echo of the original query (useful when fanout was enabled).
+      #
+      #   @return [String]
+      required :query, String
+
+      # @!attribute results
+      #
+      #   @return [Array<ContextDev::Models::WebSearchResponse::Result>]
+      required :results,
+               -> { ContextDev::Internal::Type::ArrayOf[ContextDev::Models::WebSearchResponse::Result] }
+
+      # @!method initialize(query:, results:)
+      #   @param query [String] Echo of the original query (useful when fanout was enabled).
+      #
+      #   @param results [Array<ContextDev::Models::WebSearchResponse::Result>]
+
+      class Result < ContextDev::Internal::Type::BaseModel
+        # @!attribute description
+        #   Snippet excerpt from the page.
+        #
+        #   @return [String]
+        required :description, String
+
+        # @!attribute markdown
+        #   Markdown scrape status and content for this result.
+        #
+        #   @return [ContextDev::Models::WebSearchResponse::Result::Markdown]
+        required :markdown, -> { ContextDev::Models::WebSearchResponse::Result::Markdown }
+
+        # @!attribute relevance
+        #   Model-judged relevance to the original query.
+        #
+        #   @return [Symbol, ContextDev::Models::WebSearchResponse::Result::Relevance]
+        required :relevance, enum: -> { ContextDev::Models::WebSearchResponse::Result::Relevance }
+
+        # @!attribute title
+        #   Page title.
+        #
+        #   @return [String]
+        required :title, String
+
+        # @!attribute url
+        #   Canonical result URL.
+        #
+        #   @return [String]
+        required :url, String
+
+        # @!method initialize(description:, markdown:, relevance:, title:, url:)
+        #   @param description [String] Snippet excerpt from the page.
+        #
+        #   @param markdown [ContextDev::Models::WebSearchResponse::Result::Markdown] Markdown scrape status and content for this result.
+        #
+        #   @param relevance [Symbol, ContextDev::Models::WebSearchResponse::Result::Relevance] Model-judged relevance to the original query.
+        #
+        #   @param title [String] Page title.
+        #
+        #   @param url [String] Canonical result URL.
+
+        # @see ContextDev::Models::WebSearchResponse::Result#markdown
+        class Markdown < ContextDev::Internal::Type::BaseModel
+          # @!attribute code
+          #   Per-result scrape outcome. Inspect this before reading `markdown`.
+          #
+          #   @return [Symbol, ContextDev::Models::WebSearchResponse::Result::Markdown::Code]
+          required :code, enum: -> { ContextDev::Models::WebSearchResponse::Result::Markdown::Code }
+
+          # @!attribute markdown
+          #   GFM Markdown of the page. Null unless markdownOptions.enabled is true and
+          #   scraping succeeded.
+          #
+          #   @return [String, nil]
+          required :markdown, String, nil?: true
+
+          # @!method initialize(code:, markdown:)
+          #   Some parameter documentations has been truncated, see
+          #   {ContextDev::Models::WebSearchResponse::Result::Markdown} for more details.
+          #
+          #   Markdown scrape status and content for this result.
+          #
+          #   @param code [Symbol, ContextDev::Models::WebSearchResponse::Result::Markdown::Code] Per-result scrape outcome. Inspect this before reading `markdown`.
+          #
+          #   @param markdown [String, nil] GFM Markdown of the page. Null unless markdownOptions.enabled is true and scrapi
+
+          # Per-result scrape outcome. Inspect this before reading `markdown`.
+          #
+          # @see ContextDev::Models::WebSearchResponse::Result::Markdown#code
+          module Code
+            extend ContextDev::Internal::Type::Enum
+
+            SUCCESS = :SUCCESS
+            NOT_REQUESTED = :NOT_REQUESTED
+            TIMEOUT = :TIMEOUT
+            WEBSITE_ACCESS_ERROR = :WEBSITE_ACCESS_ERROR
+            ERROR = :ERROR
+
+            # @!method self.values
+            #   @return [Array<Symbol>]
+          end
+        end
+
+        # Model-judged relevance to the original query.
+        #
+        # @see ContextDev::Models::WebSearchResponse::Result#relevance
+        module Relevance
+          extend ContextDev::Internal::Type::Enum
+
+          HIGH = :high
+          MEDIUM = :medium
+          LOW = :low
+
+          # @!method self.values
+          #   @return [Array<Symbol>]
+        end
+      end
+    end
+  end
+end

data/lib/context_dev/models.rb

@@ -73,6 +73,8 @@ module ContextDev
 
   WebScreenshotParams = ContextDev::Models::WebScreenshotParams
 
+  WebSearchParams = ContextDev::Models::WebSearchParams
+
   WebWebCrawlMdParams = ContextDev::Models::WebWebCrawlMdParams
 
   WebWebScrapeHTMLParams = ContextDev::Models::WebWebScrapeHTMLParams

data/lib/context_dev/resources/web.rb

@@ -114,6 +114,43 @@ module ContextDev
         )
       end
 
+      # Some parameter documentations has been truncated, see
+      # {ContextDev::Models::WebSearchParams} for more details.
+      #
+      # Search the web and optionally scrape each result to Markdown in one round-trip.
+      #
+      # @overload search(query:, exclude_domains: nil, freshness: nil, include_domains: nil, markdown_options: nil, query_fanout: nil, timeout_ms: nil, request_options: {})
+      #
+      # @param query [String] Natural-language search query.
+      #
+      # @param exclude_domains [Array<String>] Blocklist — drop results from these domains. Example: ["pinterest.com", "reddit.
+      #
+      # @param freshness [Symbol, ContextDev::Models::WebSearchParams::Freshness] Restrict results to content published within this window.
+      #
+      # @param include_domains [Array<String>] Allowlist — only return results from these domains. Example: ["arxiv.org", "gith
+      #
+      # @param markdown_options [ContextDev::Models::WebSearchParams::MarkdownOptions] Inline Markdown scraping for each result. Set `enabled: true` to activate.
+      #
+      # @param query_fanout [Boolean] Expand the query into multiple parallel variants for broader recall.
+      #
+      # @param timeout_ms [Integer] Optional timeout in milliseconds for the request. If the request takes longer th
+      #
+      # @param request_options [ContextDev::RequestOptions, Hash{Symbol=>Object}, nil]
+      #
+      # @return [ContextDev::Models::WebSearchResponse]
+      #
+      # @see ContextDev::Models::WebSearchParams
+      def search(params)
+        parsed, options = ContextDev::WebSearchParams.dump_request(params)
+        @client.request(
+          method: :post,
+          path: "web/search",
+          body: parsed,
+          model: ContextDev::Models::WebSearchResponse,
+          options: options
+        )
+      end
+
       # Some parameter documentations has been truncated, see
       # {ContextDev::Models::WebWebCrawlMdParams} for more details.
       #

data/lib/context_dev/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module ContextDev
-  VERSION = "1.20.1"
+  VERSION = "1.21.0"
 end

data/lib/context_dev.rb

@@ -86,6 +86,8 @@ require_relative "context_dev/models/web_extract_styleguide_params"
 require_relative "context_dev/models/web_extract_styleguide_response"
 require_relative "context_dev/models/web_screenshot_params"
 require_relative "context_dev/models/web_screenshot_response"
+require_relative "context_dev/models/web_search_params"
+require_relative "context_dev/models/web_search_response"
 require_relative "context_dev/models/web_web_crawl_md_params"
 require_relative "context_dev/models/web_web_crawl_md_response"
 require_relative "context_dev/models/web_web_scrape_html_params"