{ "openapi": "3.0.1", "info": { "title": "Company Financials Scraper — Valuation & M&A (Craft.co)", "description": "Scrape company financial intelligence from Craft.co — revenue, income statements, private valuations, market cap, M&A, executives & competitors as clean JSON. Low-cost Crunchbase / PitchBook alternative for VC/PE deal sourcing, M&A research, sales prospecting & CRM enrichment.", "version": "0.4", "x-build-id": "Nhf0Y1ikvNAG6eQWS" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/foxlabs~craft-intelligence/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-foxlabs-craft-intelligence", "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": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/foxlabs~craft-intelligence/runs": { "post": { "operationId": "runs-sync-foxlabs-craft-intelligence", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/foxlabs~craft-intelligence/run-sync": { "post": { "operationId": "run-sync-foxlabs-craft-intelligence", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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": { "preset": { "title": "Quick start: sector preset", "enum": [ "", "saas", "fintech", "ai-ml" ], "type": "string", "description": "Auto-fills well-known company anchors, then (with Discovery depth > 0) follows Craft.co's competitor graph. NOTE: the graph spans the broader tech/software space — a 'B2B Tech & Software' run also surfaces adjacent cloud, infra, dev-tools and large software vendors, not only narrow SaaS. A built-in blocklist drops pure hardware/semiconductor/telecom/electronics. Or leave blank and paste your own companies below.", "default": "" }, "companyNames": { "title": "Company names, website domains or slugs (no URL needed)", "type": "array", "description": "One company per line: a name (`Stripe`, `Procter & Gamble`, `Goldman Sachs Group`), a website domain (`stripe.com`) or a Craft.co slug (`notion-labs`). Craft.co has no public search, so each entry is turned into likely Craft.co pages (legal forms like Inc / Ltd / AG are dropped; `stripe.com` → craft.co/stripe) and a page is delivered only if it is the company you asked for: its name must match the name you typed, or its website must match the domain you gave. Entries that cannot be matched are listed in the log and in FAILED_PAGES and are never billed. ⚠ Two companies with exactly the same name cannot be told apart (`Notion` is a small smart-home company — the well-known one is `Notion Labs`), and renamed companies are not matched by their old name (`Square` → Block). When in doubt, paste the exact Craft.co URL in the field below.", "default": [], "items": { "type": "string" } }, "companyUrls": { "title": "Or paste Craft.co URLs (for tricky slugs / bulk file)", "type": "array", "description": "Craft.co URLs are the most precise input — any form works (`craft.co/stripe`, `https://www.craft.co/stripe/financials` → https://craft.co/stripe). A company website URL is treated like a domain above. You can also upload a .txt or .csv file with one company per line (Craft.co URL, name or domain; for a .csv the first column is used, and a header row is skipped).", "default": [], "items": { "type": "object", "required": [ "url" ], "properties": { "url": { "type": "string", "title": "URL of a web page", "format": "uri" } } } }, "includeCompetitors": { "title": "Include competitors", "type": "boolean", "description": "Each company's competitor list — name, Craft.co URL and revenue (competitor revenue is often null on Craft.co).", "default": true }, "includeAcquisitions": { "title": "Include M&A / acquisitions", "type": "boolean", "description": "Acquisition / subsidiary history — company, date, amount and Craft.co URL (date & amount are usually null on Craft.co).", "default": true }, "includeLeadership": { "title": "Include key executives", "type": "boolean", "description": "Executive team — name + title.", "default": true }, "maxResults": { "title": "Maximum companies", "minimum": 0, "maximum": 100000, "type": "integer", "description": "Hard cap on companies returned (also your cost cap). PPE $3.50/1K delivered + $0.50/1K looked up → 100 ≈ $0.40.", "default": 100 }, "maxConcurrency": { "title": "Concurrent fetches", "minimum": 1, "maximum": 20, "type": "integer", "description": "Pages fetched in parallel (1-20). In our test on 2026-10-06 the share of blocked requests was the same at 1, 3, 6 and 8, so this mostly sets the speed. Proxy sessions that worked are re-used; the Actor keeps as many sessions as this number.", "default": 5 }, "maxDepth": { "title": "Competitor discovery depth", "minimum": 0, "maximum": 4, "type": "integer", "description": "How far to follow Craft.co's competitor graph for a live, broad pool (not a static anchor list):\n• 0 = only the companies you list / preset anchors — no discovery\n• 1 = + each company's direct competitors\n• 2 = + competitors-of-competitors (DEFAULT for presets — densest live sweep)\n• 3 = one hop further (market-wide)\n• 4 = exhaustive crawl\nWith a preset, discovery is ON by default (depth 2) so every run reflects today's market. Set 0 to scrape only an exact company list. 'Maximum companies' is always your hard cost cap (raise it for more density).", "default": 2 }, "industryKeywords": { "title": "Industry whitelist (discovery only)", "type": "array", "description": "Only applied when depth > 0. Discovered companies whose industry tags match none of these keywords are skipped — this stops drift into unrelated sectors (e.g. fintech → e-commerce). Presets fill this automatically.", "default": [], "items": { "type": "string" } }, "industryBlocklist": { "title": "Industry blocklist (discovery only)", "type": "array", "description": "Only applied when depth > 0. Discovered companies whose industry tags match ANY of these keywords are dropped — even if they also match the whitelist. Keeps unwanted sectors (e.g. hardware, telecom, semiconductors) out of a competitor sweep. Merged with the preset's built-in blocklist.", "default": [], "items": { "type": "string" } }, "enrichFromOwler": { "title": "Backfill missing revenue/employees from Owler (estimate)", "type": "boolean", "description": "OFF by default. When ON, companies whose revenue/employees are missing on Craft.co are looked up on Owler.com and backfilled ONLY when Owler's domain matches the company's website (so a wrong-name match like 'linear' → Linear Technology is rejected, never merged). Backfilled values are tagged `revenueSource:\"owler\"` + `revenueIsEstimate:true` and a verified `owlerUrl` is added. Note: this fetches a second page per missing-data company, so it roughly doubles cost/time for those rows.", "default": false }, "proxyConfiguration": { "title": "Proxy configuration", "type": "object", "description": "RESIDENTIAL proxy is the default, with exits in the United States unless you pick a country here. Craft.co's bot protection answers part of all requests with a challenge (HTTP 403); such a request is retried on another session (up to 10 attempts for a company you asked for, 6 for a discovered competitor) and a session that worked is re-used. Measured on 2026-10-06: the datacenter proxy was blocked on 23 of 24 requests (it worked in 2 small runs on 2026-10-02), so keep RESIDENTIAL.", "default": { "useApifyProxy": true, "apifyProxyGroups": [ "RESIDENTIAL" ] } } } }, "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 } } } } } } } } } }