{ "openapi": "3.0.1", "info": { "title": "Google Academic Scraper — Scholar + Patents", "description": "One actor for Google Scholar (search, author profiles with h-index/i10/citations-by-year, cite formats) and Google Patents (search, full details, family, N-hop citation network). Pure HTTP, no login.", "version": "0.0", "x-build-id": "id2HWaUyLwXayJjM8" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/brilliant_gum~google-academic-scraper/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-brilliant_gum-google-academic-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for its completion, and returns Actor's dataset items in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/brilliant_gum~google-academic-scraper/runs": { "post": { "operationId": "runs-sync-brilliant_gum-google-academic-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/brilliant_gum~google-academic-scraper/run-sync": { "post": { "operationId": "run-sync-brilliant_gum-google-academic-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } } }, "components": { "schemas": { "inputSchema": { "type": "object", "required": [ "source" ], "properties": { "source": { "title": "Source", "enum": [ "scholar", "patents", "both" ], "type": "string", "description": "Which Google service to scrape: Google Scholar, Google Patents, or both in one run.", "default": "scholar" }, "scholarMode": { "title": "Scholar mode", "enum": [ "search", "authorProfile", "citeFormats" ], "type": "string", "description": "Scholar task: 'search' (publications), 'authorProfile' (metrics by Scholar user id), or 'citeFormats' (MLA/APA/Chicago/BibTeX for results).", "default": "search" }, "patentsMode": { "title": "Patents mode", "enum": [ "search", "details", "family", "citationNetwork", "byAssignee", "byInventor" ], "type": "string", "description": "Patents task: 'search' (free-text + filters), 'details' (full record by patent id), 'family', 'citationNetwork' (N-hop), or by assignee / by inventor.", "default": "search" }, "queries": { "title": "Queries (Scholar / both)", "type": "array", "description": "Search terms for Scholar search (one item per query). When source='both', each term is also searched on Google Patents.", "items": { "type": "string" } }, "authorIds": { "title": "Scholar author IDs", "type": "array", "description": "Google Scholar user IDs for authorProfile mode (the 'user=' value in a profile URL, e.g. WLN3QrAAAAAJ).", "items": { "type": "string" } }, "citeResultIds": { "title": "Scholar cite result IDs", "type": "array", "description": "Result data-cid / cluster ids for citeFormats mode. Leave empty to derive them automatically from 'queries'.", "items": { "type": "string" } }, "includeCiteFormats": { "title": "Include cite formats with search", "type": "boolean", "description": "When citeFormats mode derives ids from queries, also emit the citation-format record for each result.", "default": true }, "includeAnalytics": { "title": "Include derived analytics", "type": "boolean", "description": "Add computed analytics to author profiles (citation velocity, recent-2y citations, h-index cross-check, top venues).", "default": true }, "scholarYearFrom": { "title": "Scholar year from", "type": "integer", "description": "Lower bound publication year for Scholar search (Google as_ylo)." }, "scholarYearTo": { "title": "Scholar year to", "type": "integer", "description": "Upper bound publication year for Scholar search (Google as_yhi)." }, "scholarSortByDate": { "title": "Scholar sort by date", "type": "boolean", "description": "Sort Scholar search results by date instead of relevance (Google scisbd=1).", "default": false }, "scholarPatentsCourts": { "title": "Scholar scope", "enum": [ "0,5", "7" ], "type": "string", "description": "Scholar result scope (Google as_sdt): articles only, or include patents & case law.", "default": "0,5" }, "includeCitations": { "title": "Include citations-only entries", "type": "boolean", "description": "Include Scholar entries that are citations only (no full record). Uncheck to hide them (as_vis=1).", "default": true }, "includeUnfilteredResults": { "title": "Include unfiltered (omitted) results", "type": "boolean", "description": "Append filter=0 to include near-duplicate/omitted Scholar results.", "default": false }, "noAutocorrect": { "title": "No spelling autocorrect", "type": "boolean", "description": "Disable Scholar spelling auto-correction (Google nfpr=1).", "default": false }, "languageResults": { "title": "Results language (lr)", "type": "string", "description": "Restrict Scholar results to a language, e.g. lang_en (Google lr)." }, "countryResults": { "title": "Results country (cr)", "type": "string", "description": "Restrict Scholar results to a country, e.g. countryUS (Google cr)." }, "searchQuery": { "title": "Patents search query", "type": "string", "description": "Free-text query for Google Patents search (e.g. 'neural network'). Alias of queries[0] when source='patents'." }, "patentIds": { "title": "Patent IDs", "type": "array", "description": "Patent numbers for details / family / citationNetwork modes, e.g. US10000000B2.", "items": { "type": "string" } }, "assignee": { "title": "Assignee filter", "type": "string", "description": "Filter Google Patents by assignee / owner name (e.g. 'Apple Inc.'). Drives byAssignee mode." }, "inventor": { "title": "Inventor filter", "type": "string", "description": "Filter Google Patents by inventor name (e.g. 'Geoffrey Hinton'). Drives byInventor mode." }, "countryCodes": { "title": "Patent country codes", "type": "array", "description": "Restrict Google Patents results to these country/office codes, e.g. US, EP, WO, CN.", "items": { "type": "string" } }, "status": { "title": "Patent status", "enum": [ "ANY", "GRANT", "APPLICATION" ], "type": "string", "description": "Filter by grant status.", "default": "ANY" }, "patentType": { "title": "Patent type", "enum": [ "ANY", "PATENT", "DESIGN" ], "type": "string", "description": "Filter by patent type.", "default": "ANY" }, "dateFrom": { "title": "Date from (YYYY-MM-DD)", "type": "string", "description": "Lower date bound for Google Patents search (applies to the date type below)." }, "dateTo": { "title": "Date to (YYYY-MM-DD)", "type": "string", "description": "Upper date bound for Google Patents search (applies to the date type below)." }, "dateType": { "title": "Patent date type", "enum": [ "priority", "filing", "publication" ], "type": "string", "description": "Which date the dateFrom/dateTo range filters on.", "default": "priority" }, "patentSort": { "title": "Patent sort", "enum": [ "relevance", "new", "old" ], "type": "string", "description": "Sort order for Google Patents search results.", "default": "relevance" }, "patentLanguage": { "title": "Patent language", "type": "string", "description": "Restrict Google Patents results to a language (e.g. ENGLISH)." }, "enrichmentDepth": { "title": "Patent enrichment depth", "enum": [ "basic", "deep" ], "type": "string", "description": "'basic' returns search fields only; 'deep' additionally fetches each patent's detail page for claims, CPC/IPC, citations and family.", "default": "basic" }, "includeFamily": { "title": "Include patent family", "type": "boolean", "description": "Include worldwide application / family members in patent detail records.", "default": true }, "includeClaims": { "title": "Include full claim text", "type": "boolean", "description": "Include full claim text in patent detail records (large). Claims count is always included.", "default": true }, "citationDirection": { "title": "Citation network direction", "enum": [ "backward", "forward", "both" ], "type": "string", "description": "Which citations to expand in citationNetwork mode.", "default": "both" }, "citationDepth": { "title": "Citation network depth", "minimum": 1, "maximum": 3, "type": "integer", "description": "How many hops to expand in citationNetwork mode (1-3). Nodes are globally de-duplicated across hops.", "default": 1 }, "maxResults": { "title": "Max results", "minimum": 1, "maximum": 5000, "type": "integer", "description": "Maximum results per query (Scholar) / per search query (Patents) / node cap (citation network).", "default": 20 }, "maxPages": { "title": "Max pages (optional)", "minimum": 1, "maximum": 100, "type": "integer", "description": "Explicit page cap. If empty, derived from Max results (Scholar: 10/page, Patents: 100/page)." }, "countryCode": { "title": "Country (gl)", "type": "string", "description": "Country of search for Scholar and proxy geolocation (Google gl, e.g. us).", "default": "us" }, "languageCode": { "title": "Language (hl)", "type": "string", "description": "Interface language for Scholar/Patents (Google hl, e.g. en).", "default": "en" }, "proxyConfiguration": { "title": "Proxy configuration", "type": "object", "description": "Proxy configuration. Leave as-is — proxies are tuned automatically for reliable results. Requires an Apify plan with proxy access.", "default": { "useApifyProxy": true } } } }, "runsResponseSchema": { "type": "object", "properties": { "data": { "type": "object", "properties": { "id": { "type": "string" }, "actId": { "type": "string" }, "userId": { "type": "string" }, "startedAt": { "type": "string", "format": "date-time", "example": "2025-01-08T00:00:00.000Z" }, "finishedAt": { "type": "string", "format": "date-time", "example": "2025-01-08T00:00:00.000Z" }, "status": { "type": "string", "example": "READY" }, "meta": { "type": "object", "properties": { "origin": { "type": "string", "example": "API" }, "userAgent": { "type": "string" } } }, "stats": { "type": "object", "properties": { "inputBodyLen": { "type": "integer", "example": 2000 }, "rebootCount": { "type": "integer", "example": 0 }, "restartCount": { "type": "integer", "example": 0 }, "resurrectCount": { "type": "integer", "example": 0 }, "computeUnits": { "type": "integer", "example": 0 } } }, "options": { "type": "object", "properties": { "build": { "type": "string", "example": "latest" }, "timeoutSecs": { "type": "integer", "example": 300 }, "memoryMbytes": { "type": "integer", "example": 1024 }, "diskMbytes": { "type": "integer", "example": 2048 } } }, "buildId": { "type": "string" }, "defaultKeyValueStoreId": { "type": "string" }, "defaultDatasetId": { "type": "string" }, "defaultRequestQueueId": { "type": "string" }, "buildNumber": { "type": "string", "example": "1.0.0" }, "containerUrl": { "type": "string" }, "usage": { "type": "object", "properties": { "ACTOR_COMPUTE_UNITS": { "type": "integer", "example": 0 }, "DATASET_READS": { "type": "integer", "example": 0 }, "DATASET_WRITES": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_READS": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_WRITES": { "type": "integer", "example": 1 }, "KEY_VALUE_STORE_LISTS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_READS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_WRITES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_INTERNAL_GBYTES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_EXTERNAL_GBYTES": { "type": "integer", "example": 0 }, "PROXY_RESIDENTIAL_TRANSFER_GBYTES": { "type": "integer", "example": 0 }, "PROXY_SERPS": { "type": "integer", "example": 0 }, "PROXY_UNBLOCKER_UNITS": { "type": "integer", "example": 0 } } }, "usageTotalUsd": { "type": "number", "example": 0.00005 }, "usageUsd": { "type": "object", "properties": { "ACTOR_COMPUTE_UNITS": { "type": "integer", "example": 0 }, "DATASET_READS": { "type": "integer", "example": 0 }, "DATASET_WRITES": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_READS": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_WRITES": { "type": "number", "example": 0.00005 }, "KEY_VALUE_STORE_LISTS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_READS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_WRITES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_INTERNAL_GBYTES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_EXTERNAL_GBYTES": { "type": "integer", "example": 0 }, "PROXY_RESIDENTIAL_TRANSFER_GBYTES": { "type": "integer", "example": 0 }, "PROXY_SERPS": { "type": "integer", "example": 0 }, "PROXY_UNBLOCKER_UNITS": { "type": "integer", "example": 0 } } } } } } } } } }