context.dev 2.8.0 → 2.10.0

data/rbi/context_dev/models/web_web_scrape_md_response.rbi

@@ -72,6 +72,15 @@ module ContextDev
       sig { params(actions_html_stale: T::Boolean).void }
       attr_writer :actions_html_stale
 
+      # Only present when includeHTML=true: the page HTML the Markdown was converted
+      # from — the same body the Scrape HTML endpoint returns for the equivalent
+      # request.
+      sig { returns(T.nilable(String)) }
+      attr_reader :html
+
+      sig { params(html: String).void }
+      attr_writer :html
+
       # Metadata about the API key used for the request. Included in every response
       # whenever a valid API key is provided, even when the response status is not 200.
       sig do
@@ -103,6 +112,7 @@ module ContextDev
               ContextDev::Models::WebWebScrapeMdResponse::ActionsApplied::OrHash
             ],
           actions_html_stale: T::Boolean,
+          html: String,
           key_metadata:
             ContextDev::Models::WebWebScrapeMdResponse::KeyMetadata::OrHash
         ).returns(T.attached_class)
@@ -125,6 +135,10 @@ module ContextDev
         # True when an action was applied but the returned content could not be refreshed
         # afterward.
         actions_html_stale: nil,
+        # Only present when includeHTML=true: the page HTML the Markdown was converted
+        # from — the same body the Scrape HTML endpoint returns for the equivalent
+        # request.
+        html: nil,
         # Metadata about the API key used for the request. Included in every response
         # whenever a valid API key is provided, even when the response status is not 200.
         key_metadata: nil
@@ -145,6 +159,7 @@ module ContextDev
                 ContextDev::Models::WebWebScrapeMdResponse::ActionsApplied
               ],
             actions_html_stale: T::Boolean,
+            html: String,
             key_metadata:
               ContextDev::Models::WebWebScrapeMdResponse::KeyMetadata
           }
@@ -245,6 +260,29 @@ module ContextDev
         sig { params(favicon: String).void }
         attr_writer :favicon
 
+        # Page headings (h1–h6) in document order, extracted from the unfiltered document.
+        # Capped at the first 500 headings. Omitted when the page has none.
+        sig do
+          returns(
+            T.nilable(
+              T::Array[
+                ContextDev::Models::WebWebScrapeMdResponse::Metadata::Heading
+              ]
+            )
+          )
+        end
+        attr_reader :headings
+
+        sig do
+          params(
+            headings:
+              T::Array[
+                ContextDev::Models::WebWebScrapeMdResponse::Metadata::Heading::OrHash
+              ]
+          ).void
+        end
+        attr_writer :headings
+
         # Primary resolved preview image from Open Graph, Twitter, or image metadata.
         sig { returns(T.nilable(String)) }
         attr_reader :image
@@ -374,6 +412,10 @@ module ContextDev
             canonical_url: String,
             description: String,
             favicon: String,
+            headings:
+              T::Array[
+                ContextDev::Models::WebWebScrapeMdResponse::Metadata::Heading::OrHash
+              ],
             image: String,
             json_ld: T::Array[T::Hash[Symbol, T.anything]],
             keywords: T::Array[String],
@@ -413,6 +455,9 @@ module ContextDev
           description: nil,
           # Resolved favicon URL, when present.
           favicon: nil,
+          # Page headings (h1–h6) in document order, extracted from the unfiltered document.
+          # Capped at the first 500 headings. Omitted when the page has none.
+          headings: nil,
           # Primary resolved preview image from Open Graph, Twitter, or image metadata.
           image: nil,
           # JSON-LD structured data blocks parsed from the page.
@@ -456,6 +501,10 @@ module ContextDev
               canonical_url: String,
               description: String,
               favicon: String,
+              headings:
+                T::Array[
+                  ContextDev::Models::WebWebScrapeMdResponse::Metadata::Heading
+                ],
               image: String,
               json_ld: T::Array[T::Hash[Symbol, T.anything]],
               keywords: T::Array[String],
@@ -566,6 +615,37 @@ module ContextDev
           end
         end
 
+        class Heading < ContextDev::Internal::Type::BaseModel
+          OrHash =
+            T.type_alias do
+              T.any(
+                ContextDev::Models::WebWebScrapeMdResponse::Metadata::Heading,
+                ContextDev::Internal::AnyHash
+              )
+            end
+
+          # Heading level, 1–6 (from h1–h6).
+          sig { returns(Integer) }
+          attr_accessor :level
+
+          # Heading text with whitespace collapsed, truncated to 1000 characters.
+          sig { returns(String) }
+          attr_accessor :text
+
+          sig { params(level: Integer, text: String).returns(T.attached_class) }
+          def self.new(
+            # Heading level, 1–6 (from h1–h6).
+            level:,
+            # Heading text with whitespace collapsed, truncated to 1000 characters.
+            text:
+          )
+          end
+
+          sig { override.returns({ level: Integer, text: String }) }
+          def to_hash
+          end
+        end
+
         module OpenGraph
           extend ContextDev::Internal::Type::Union
 

data/rbi/context_dev/models/web_web_scrape_sitemap_params.rbi

@@ -35,6 +35,15 @@ module ContextDev
       sig { params(max_links: Integer).void }
       attr_writer :max_links
 
+      # Optional search phrase. When provided, the crawled sitemap is filtered to the
+      # pages whose URLs are about that phrase, most relevant first, and the request
+      # costs 2 credits instead of 1.
+      sig { returns(T.nilable(String)) }
+      attr_reader :search
+
+      sig { params(search: String).void }
+      attr_writer :search
+
       # Optional explicit sitemap URL. When provided, exactly this sitemap is crawled
       # instead of discovering the domain's sitemaps.
       sig { returns(T.nilable(String)) }
@@ -88,6 +97,7 @@ module ContextDev
           domain: String,
           headers: T::Hash[Symbol, String],
           max_links: Integer,
+          search: String,
           sitemap_url: String,
           tags: T::Array[String],
           timeout_ms: Integer,
@@ -106,6 +116,10 @@ module ContextDev
         # 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 search phrase. When provided, the crawled sitemap is filtered to the
+        # pages whose URLs are about that phrase, most relevant first, and the request
+        # costs 2 credits instead of 1.
+        search: nil,
         # Optional explicit sitemap URL. When provided, exactly this sitemap is crawled
         # instead of discovering the domain's sitemaps.
         sitemap_url: nil,
@@ -135,6 +149,7 @@ module ContextDev
             domain: String,
             headers: T::Hash[Symbol, String],
             max_links: Integer,
+            search: String,
             sitemap_url: String,
             tags: T::Array[String],
             timeout_ms: Integer,

data/rbi/context_dev/models/web_web_scrape_sitemap_response.rbi

@@ -34,7 +34,8 @@ module ContextDev
       end
       attr_accessor :success
 
-      # Array of discovered page URLs from the sitemap (max 500)
+      # Discovered page URLs from the sitemap, up to `maxLinks`. When `search` is set
+      # these are only the matching pages, most relevant first.
       sig { returns(T::Array[String]) }
       attr_accessor :urls
 
@@ -75,7 +76,8 @@ module ContextDev
         meta:,
         # Indicates success
         success:,
-        # Array of discovered page URLs from the sitemap (max 500)
+        # Discovered page URLs from the sitemap, up to `maxLinks`. When `search` is set
+        # these are only the matching pages, most relevant first.
         urls:,
         # Metadata about the API key used for the request. Included in every response
         # whenever a valid API key is provided, even when the response status is not 200.

data/rbi/context_dev/models.rbi

@@ -7,6 +7,8 @@ module ContextDev
 
   BatchCancelParams = ContextDev::Models::BatchCancelParams
 
+  BatchDeleteParams = ContextDev::Models::BatchDeleteParams
+
   BatchGetResultsParams = ContextDev::Models::BatchGetResultsParams
 
   BatchListParams = ContextDev::Models::BatchListParams
@@ -60,10 +62,14 @@ module ContextDev
 
   MonitorUpdateParams = ContextDev::Models::MonitorUpdateParams
 
+  NewsSearchParams = ContextDev::Models::NewsSearchParams
+
   PageErrorCount = ContextDev::Models::PageErrorCount
 
   ParseHandleParams = ContextDev::Models::ParseHandleParams
 
+  PersonEnrichParams = ContextDev::Models::PersonEnrichParams
+
   UtilityPrefetchParams = ContextDev::Models::UtilityPrefetchParams
 
   WebExtractCompetitorsParams = ContextDev::Models::WebExtractCompetitorsParams

data/rbi/context_dev/resources/batch.rbi

@@ -2,6 +2,7 @@
 
 module ContextDev
   module Resources
+    # Scrape many pages or crawl a site asynchronously.
     class Batch
       # Check progress, and get download links once the batch finishes.
       sig do
@@ -49,6 +50,21 @@ module ContextDev
       )
       end
 
+      # Permanently delete a finished batch and its stored results. Active batches must
+      # settle first.
+      sig do
+        params(
+          batch_id: String,
+          request_options: ContextDev::RequestOptions::OrHash
+        ).returns(ContextDev::Models::BatchDeleteResponse)
+      end
+      def delete(
+        # ID of the batch to retrieve or cancel.
+        batch_id,
+        request_options: {}
+      )
+      end
+
       # Stop a batch from starting new pages. In-progress pages finish, and unused
       # credits are refunded.
       sig do
@@ -86,24 +102,30 @@ module ContextDev
       )
       end
 
-      # Retrieve and normalize a person profile from identifiers.
+      # Scrape 25K URLs or crawl large websites asynchronously.
       sig do
         params(
-          identifiers: ContextDev::BatchSubmitParams::Identifiers::OrHash,
+          input:
+            T.any(
+              ContextDev::BatchSubmitParams::Input::Scrape::OrHash,
+              ContextDev::BatchSubmitParams::Input::Crawl::OrHash
+            ),
           tags: T::Array[String],
-          timeout_ms: Integer,
+          webhook_url: String,
+          idempotency_key: String,
           request_options: ContextDev::RequestOptions::OrHash
         ).returns(ContextDev::Models::BatchSubmitResponse)
       end
       def submit(
-        # Known identifiers for the person. At least one identifier is required.
-        identifiers:,
-        # Optional tags for tracking usage. Up to 20 tags, each 1 to 50 characters.
+        # Body param: Choose a URL list or a site crawl.
+        input:,
+        # Body param: Tags stored on the batch. Filter the batch list by them later.
         tags: 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,
+        # Body param: URL notified when the batch finishes.
+        webhook_url: nil,
+        # Header param: Any string unique to this submission. Retries with the same key
+        # return the original batch.
+        idempotency_key: nil,
         request_options: {}
       )
       end

data/rbi/context_dev/resources/brand.rbi

@@ -64,28 +64,34 @@ module ContextDev
       )
       end
 
-      # Search brands by name or domain and get back up to 10 lightweight matches
-      # (domain, name, logo), most popular first: by Tranco rank, then market cap for
-      # brands outside the Tranco list, with text relevance breaking ties. Matching is
-      # prefix-based with no typo tolerance, so it is suited to autocomplete. Only
-      # brands already in the Context.dev index are returned — use /brand/retrieve to
-      # fetch (and index) a specific domain. Free on Pro and Scale plans; costs 1 credit
-      # per request on the Free and Starter plans.
+      # Search indexed brands by name or domain
       sig do
         params(
           query: String,
+          autocomplete: T::Boolean,
+          query_by: T::Array[ContextDev::BrandSearchParams::QueryBy::OrSymbol],
           tags: T::Array[String],
+          typo_tolerance: Integer,
           request_options: ContextDev::RequestOptions::OrHash
         ).returns(ContextDev::Models::BrandSearchResponse)
       end
       def search(
-        # Search term, matched against brand names and domains by prefix (e.g. 'nike',
+        # Search term, matched against the fields selected by queryBy (e.g. 'nike',
         # 'nike.com', 'nik').
         query:,
+        # Whether the search term matches by prefix, so partial words match as they are
+        # typed (e.g. 'nik' matches Nike). Set to false to match whole words only.
+        autocomplete: nil,
+        # Fields to match the search term against, as a comma-separated list or repeated
+        # parameter: 'name', 'domain', or both. Defaults to both.
+        query_by: nil,
         # Optional comma-separated caller-defined tags for tracking this request. Tags are
         # recorded on the request's usage log and can be used to filter usage on the
         # dashboard usage page. Up to 20 tags, each 1-50 characters.
         tags: nil,
+        # Maximum number of typos tolerated when matching, from 0 to 2. Defaults to 0 (no
+        # typo tolerance).
+        typo_tolerance: nil,
         request_options: {}
       )
       end

data/rbi/context_dev/resources/news.rbi

@@ -0,0 +1,46 @@
+# typed: strong
+
+module ContextDev
+  module Resources
+    # Search live first-party RSS and free historical news data by company identity.
+    class News
+      # Searches live and historical company news for one company, identified in
+      # searchBy by name, domain, ticker (optionally disambiguated by exchange), or
+      # ISIN. Results can be filtered by publisher domain, publisher country, article
+      # language, article type, and published-at date, and include stable story IDs,
+      # source metadata, verified entity relevance, and cursor pagination.
+      sig do
+        params(
+          search_by: ContextDev::NewsSearchParams::SearchBy::OrHash,
+          cursor: T.nilable(String),
+          filter_by: ContextDev::NewsSearchParams::FilterBy::OrHash,
+          limit: Integer,
+          sort_by: ContextDev::NewsSearchParams::SortBy::OrHash,
+          tags: T::Array[String],
+          request_options: ContextDev::RequestOptions::OrHash
+        ).returns(ContextDev::Models::NewsSearchResponse)
+      end
+      def search(
+        # What to search for.
+        search_by:,
+        # Opaque next_cursor from the previous response, or null for the first page.
+        cursor: nil,
+        # Optional result filters.
+        filter_by: nil,
+        # Maximum results to return. Defaults to 10.
+        limit: nil,
+        # Result ordering. Defaults to newest.
+        sort_by: nil,
+        # Optional tags for tracking usage. Up to 20 tags, each 1 to 50 characters.
+        tags: nil,
+        request_options: {}
+      )
+      end
+
+      # @api private
+      sig { params(client: ContextDev::Client).returns(T.attached_class) }
+      def self.new(client:)
+      end
+    end
+  end
+end

data/rbi/context_dev/resources/parse.rbi

@@ -10,29 +10,13 @@ module ContextDev
           body: ContextDev::Internal::FileInput,
           client: String,
           extension: ContextDev::ParseHandleParams::Extension::OrSymbol,
-          include_images:
-            T.any(
-              T::Boolean,
-              ContextDev::ParseHandleParams::IncludeImages::OrSymbol
-            ),
-          include_links:
-            T.any(
-              T::Boolean,
-              ContextDev::ParseHandleParams::IncludeLinks::OrSymbol
-            ),
-          ocr: T.any(T::Boolean, ContextDev::ParseHandleParams::Ocr::OrSymbol),
+          include_images: T::Boolean,
+          include_links: T::Boolean,
+          ocr: T::Boolean,
           pdf: ContextDev::ParseHandleParams::Pdf::OrHash,
-          shorten_base64_images:
-            T.any(
-              T::Boolean,
-              ContextDev::ParseHandleParams::ShortenBase64Images::OrSymbol
-            ),
+          shorten_base64_images: T::Boolean,
           tags: T::Array[String],
-          use_main_content_only:
-            T.any(
-              T::Boolean,
-              ContextDev::ParseHandleParams::UseMainContentOnly::OrSymbol
-            ),
+          use_main_content_only: T::Boolean,
           zdr: ContextDev::ParseHandleParams::Zdr::OrSymbol,
           request_options: ContextDev::RequestOptions::OrHash
         ).returns(ContextDev::Models::ParseHandleResponse)
@@ -49,11 +33,11 @@ module ContextDev
         include_images: nil,
         # Query param: Preserve hyperlinks in Markdown output
         include_links: nil,
-        # Query param: When true for PDF inputs, detect and OCR images embedded in the
-        # selected pages, inserting recognized text at each image's position in page
-        # reading order while preserving the PDF text layer. pdf.start/pdf.end limit the
-        # inclusive page range. When false, all OCR is disabled, including the automatic
-        # scanned-PDF fallback.
+        # Query param: When true for PDF inputs, OCR the selected pages that have no
+        # usable text layer (scans), replacing each recovered page's text with the OCR
+        # result while pages with a real text layer keep it. pdf.start/pdf.end limit the
+        # inclusive page range. Billed at 1 credit per page OCR actually recovered, on top
+        # of the base request cost. When false, no OCR runs.
         ocr: nil,
         # Query param: PDF page-range options as a JSON object, e.g. {"start": 2, "end":
         # 5}.

data/rbi/context_dev/resources/people.rbi

@@ -0,0 +1,47 @@
+# typed: strong
+
+module ContextDev
+  module Resources
+    class People
+      # Finds and normalizes the best available person candidate from additive identity
+      # clues, then assigns an identity match score from 0 to 100. Available on all paid
+      # plans. Successful requests cost 20 credits. Disposable and free email addresses
+      # (like gmail.com, yahoo.com) will throw a 422 error.
+      sig do
+        params(
+          company: ContextDev::PersonEnrichParams::Company::OrHash,
+          education:
+            T::Array[ContextDev::PersonEnrichParams::Education::OrHash],
+          email: String,
+          location: ContextDev::PersonEnrichParams::Location::OrHash,
+          name: ContextDev::PersonEnrichParams::Name::OrHash,
+          social_urls: T::Array[String],
+          tags: T::Array[String],
+          timeout_ms: Integer,
+          request_options: ContextDev::RequestOptions::OrHash
+        ).returns(ContextDev::Models::PersonEnrichResponse)
+      end
+      def enrich(
+        company: nil,
+        education: nil,
+        email: nil,
+        location: nil,
+        name: nil,
+        social_urls: nil,
+        # Optional tags for tracking usage. Up to 20 tags, each 1 to 50 characters.
+        tags: 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
+
+      # @api private
+      sig { params(client: ContextDev::Client).returns(T.attached_class) }
+      def self.new(client:)
+      end
+    end
+  end
+end

data/rbi/context_dev/resources/utility.rbi

@@ -3,10 +3,11 @@
 module ContextDev
   module Resources
     class Utility
-      # Signal that you may fetch brand data soon to improve latency. The type field
-      # selects what to prefetch (currently only 'brand') and identifier carries exactly
-      # one lookup key: a domain, or an email whose domain is extracted and validated
-      # (free email providers and disposable email addresses are not allowed).
+      # Signal that you may fetch data soon to improve latency. The type field selects
+      # what to prefetch ('brand' queues a brand data fetch, 'styleguide' queues a
+      # styleguide extraction) and identifier carries exactly one lookup key: a domain,
+      # or an email whose domain is extracted and validated (free email providers and
+      # disposable email addresses are not allowed).
       sig do
         params(
           identifier:
@@ -21,9 +22,10 @@ module ContextDev
         ).returns(ContextDev::Models::UtilityPrefetchResponse)
       end
       def prefetch(
-        # Identifier of the brand to prefetch. Provide exactly one of domain or email.
+        # Identifier of the target to prefetch. Provide exactly one of domain or email.
         identifier:,
-        # What to prefetch. Currently only 'brand' is supported.
+        # What to prefetch: 'brand' warms the brand data cache, 'styleguide' warms the
+        # styleguide cache.
         type:,
         # Optional tags for tracking usage. Up to 20 tags, each 1 to 50 characters.
         tags: nil,