{ "openapi": "3.0.1", "info": { "title": "Crunchbase Company Scraper - Funding, People & B2B Intel", "description": "Scrape Crunchbase company profiles with funding rounds, key people, and B2B intel. Build investor and startup databases at scale.", "version": "0.1", "x-build-id": "XCqSfhnGlMDeJBLS1" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/george.the.developer~unblockable-crunchbase-scraper-b2b-intelligence-leads/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-george.the.developer-unblockable-crunchbase-scraper-b2b-intelligence-leads", "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": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/george.the.developer~unblockable-crunchbase-scraper-b2b-intelligence-leads/runs": { "post": { "operationId": "runs-sync-george.the.developer-unblockable-crunchbase-scraper-b2b-intelligence-leads", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/george.the.developer~unblockable-crunchbase-scraper-b2b-intelligence-leads/run-sync": { "post": { "operationId": "run-sync-george.the.developer-unblockable-crunchbase-scraper-b2b-intelligence-leads", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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", "properties": { "searchUrls": { "title": "Crunchbase Search URLs", "type": "array", "description": "Crunchbase search/discover URLs to scrape companies from.", "items": { "type": "string" }, "default": [] }, "companyUrls": { "title": "Company Profile URLs", "type": "array", "description": "Direct Crunchbase organization profile URLs.", "items": { "type": "string" }, "default": [] }, "keywords": { "title": "Search Keywords", "type": "array", "description": "Best-effort keyword searches using Crunchbase search pages.", "items": { "type": "string" }, "default": [] }, "maxCompanies": { "title": "Maximum Companies", "minimum": 1, "maximum": 10000, "type": "integer", "description": "Maximum number of company profiles to extract.", "default": 100 }, "includeFunding": { "title": "Include Funding Data", "type": "boolean", "description": "Extract funding rounds and funding summary fields when available.", "default": true }, "includePeople": { "title": "Include Key People", "type": "boolean", "description": "Extract founders and key people data when available.", "default": true }, "includeInvestors": { "title": "Include Investors", "type": "boolean", "description": "Extract investor lists and investor-related fields when available.", "default": true }, "deltaOnly": { "title": "Delta Only", "type": "boolean", "description": "Return only companies not seen in the previous run state.", "default": false }, "outputFormat": { "title": "Output Format", "enum": [ "json", "csv", "excel" ], "type": "string", "description": "Primary output artifact written to the default key-value store OUTPUT key.", "default": "json" }, "maxConcurrency": { "title": "Max Concurrency", "minimum": 1, "maximum": 20, "type": "integer", "description": "Maximum parallel browser pages used by the crawler.", "default": 2 }, "minDelayMs": { "title": "Minimum Delay (ms)", "minimum": 0, "maximum": 60000, "type": "integer", "description": "Minimum randomized delay before each navigation.", "default": 2000 }, "maxDelayMs": { "title": "Maximum Delay (ms)", "minimum": 0, "maximum": 120000, "type": "integer", "description": "Maximum randomized delay before each navigation.", "default": 5000 }, "maxSearchPagination": { "title": "Search Pagination Depth", "minimum": 0, "maximum": 50, "type": "integer", "description": "Maximum search/discover pages to follow per seed URL.", "default": 3 }, "requestTimeoutSecs": { "title": "Request Timeout (seconds)", "minimum": 10, "maximum": 300, "type": "integer", "description": "Navigation timeout for each page request.", "default": 60 }, "ngStateWaitSecs": { "title": "ng-state Wait (seconds)", "minimum": 0, "maximum": 120, "type": "integer", "description": "Additional time to wait for Crunchbase to populate hidden app state scripts before parsing.", "default": 12 }, "warmupHomepage": { "title": "Warm Up Homepage", "type": "boolean", "description": "Open Crunchbase homepage once per session before target pages to seed cookies and reduce immediate challenges.", "default": true }, "challengeHoldSecs": { "title": "Challenge Hold (seconds)", "minimum": 0, "maximum": 120, "type": "integer", "description": "Time to wait and recheck when a Cloudflare challenge is detected before failing the request.", "default": 20 }, "challengeReloads": { "title": "Challenge Reloads", "minimum": 0, "maximum": 5, "type": "integer", "description": "Number of reload attempts after a Cloudflare challenge is detected.", "default": 1 }, "enableStealthPatches": { "title": "Enable Stealth Patches", "type": "boolean", "description": "Inject lightweight page patches (webdriver/chrome runtime/permissions) on every browser page.", "default": true }, "humanizeInteractions": { "title": "Humanize Interactions", "type": "boolean", "description": "Perform small mouse/scroll interactions during warmup and challenge recovery.", "default": true }, "capSolver": { "title": "CapSolver (Turnstile)", "type": "object", "description": "Optional CapSolver integration for Cloudflare Turnstile challenges. Prefer setting API key as an Apify secret/env var (CAPSOLVER_API_KEY).", "properties": { "enabled": { "title": "Enabled", "type": "boolean", "description": "Enable CapSolver challenge-solving attempts before fallback/failure.", "default": false }, "apiKey": { "title": "API Key", "type": "string", "description": "CapSolver API key (not recommended in input; use an Apify secret env var instead).", "default": "" }, "apiKeyEnvVar": { "title": "API Key Env Var", "type": "string", "description": "Environment variable name that contains the CapSolver API key.", "default": "CAPSOLVER_API_KEY" }, "solveTurnstile": { "title": "Solve Turnstile", "type": "boolean", "description": "Attempt to solve Cloudflare Turnstile widgets/challenge pages with CapSolver.", "default": true }, "timeoutSecs": { "title": "Solve Timeout (seconds)", "type": "integer", "description": "Maximum time to wait for CapSolver task completion.", "default": 120, "minimum": 10, "maximum": 300 }, "pollIntervalMs": { "title": "Poll Interval (ms)", "type": "integer", "description": "Polling interval for CapSolver task status.", "default": 1200, "minimum": 500, "maximum": 10000 }, "applyTokenTimeoutSecs": { "title": "Post-token Wait (seconds)", "type": "integer", "description": "Time to wait after injecting the solved Turnstile token for challenge clearance.", "default": 30, "minimum": 5, "maximum": 120 } }, "default": { "enabled": false, "apiKeyEnvVar": "CAPSOLVER_API_KEY", "solveTurnstile": true } }, "managedFallback": { "title": "Managed Fallback", "type": "object", "description": "Optional external anti-bot fallback used only when the normal browser path is challenged or cannot find ng-state.", "properties": { "enabled": { "title": "Enabled", "type": "boolean", "description": "Enable managed anti-bot fallback for company pages.", "default": false }, "mode": { "title": "Mode", "type": "string", "description": "Fallback transport mode. Use 'http' for unblocker APIs that return HTML, or 'cdp' for managed browser endpoints.", "enum": [ "http", "cdp" ], "default": "http" }, "timeoutSecs": { "title": "Timeout (seconds)", "type": "integer", "description": "Fallback request/browser timeout.", "default": 90, "minimum": 10, "maximum": 300 }, "waitAfterLoadSecs": { "title": "Wait After Load (seconds)", "type": "integer", "description": "Extra wait after navigation in fallback mode before parsing page HTML.", "default": 6, "minimum": 0, "maximum": 60 }, "ngStateWaitSecs": { "title": "Fallback ng-state Wait (seconds)", "type": "integer", "description": "How long the fallback browser should poll for ng-state.", "default": 20, "minimum": 0, "maximum": 120 }, "httpUrlTemplate": { "title": "HTTP URL Template", "type": "string", "description": "URL template for an unblocker API that returns HTML. Must include {url} placeholder, e.g. ZenRows/Scrapfly URL template.", "default": "" }, "httpMethod": { "title": "HTTP Method", "type": "string", "description": "HTTP method for the unblocker endpoint.", "enum": [ "GET", "POST" ], "default": "GET" }, "httpHeaders": { "title": "HTTP Headers", "type": "object", "description": "Optional headers for the unblocker API request (for API keys, content type, etc.).", "default": {} }, "httpBodyTemplate": { "title": "HTTP Body Template", "type": "string", "description": "Optional request body template (supports {url}) for POST-based unblocker APIs.", "default": "" }, "httpResponseIsJson": { "title": "HTTP Response Is JSON", "type": "boolean", "description": "Set true if the unblocker API returns JSON and HTML is nested in a field.", "default": false }, "httpHtmlJsonPath": { "title": "HTTP HTML JSON Path", "type": "string", "description": "Dot-path to HTML inside JSON responses (for example 'html' or 'data.html').", "default": "html" }, "cdpEndpoint": { "title": "CDP Endpoint", "type": "string", "description": "Remote managed browser CDP endpoint (wss://... or http(s)://...) used in 'cdp' mode.", "default": "" }, "cdpHeaders": { "title": "CDP Headers", "type": "object", "description": "Optional headers for CDP connection if required by provider.", "default": {} } }, "default": { "enabled": false, "mode": "http" } }, "proxyConfiguration": { "title": "Proxy Configuration", "type": "object", "description": "Apify proxy settings. Use residential proxies for best results." } } }, "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 } } }, "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 } } } } } } } } } }