{ "openapi": "3.0.1", "info": { "title": "TheCrawler — validated extraction contracts for AI agents", "description": "Validated extraction contracts for AI agents: crawl pages, diagnose readiness before LLM calls, and extract typed JSON with validation evidence. Markdown, structured errors, PDF/DOCX.", "version": "0.1", "x-build-id": "edDDhY8gx5z5fR1La" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/accurate_pouch~the-crawler/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-accurate_pouch-the-crawler", "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": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/accurate_pouch~the-crawler/runs": { "post": { "operationId": "runs-sync-accurate_pouch-the-crawler", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/accurate_pouch~the-crawler/run-sync": { "post": { "operationId": "run-sync-accurate_pouch-the-crawler", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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": [ "urls" ], "properties": { "urls": { "title": "URLs to Scrape", "type": "array", "description": "List of URLs to scrape.", "items": { "type": "string" } }, "extractText": { "title": "Extract Text", "type": "boolean", "description": "Extract visible text content (scripts/styles removed).", "default": true }, "extractLinks": { "title": "Extract Links", "type": "boolean", "description": "Extract all links with href, anchor text, and internal/external flag.", "default": true }, "extractImages": { "title": "Extract Images", "type": "boolean", "description": "Extract all images with src, alt text, dimensions.", "default": true }, "extractMeta": { "title": "Extract Meta Tags", "type": "boolean", "description": "Extract all meta tags (description, og:title, keywords, etc).", "default": true }, "extractHeadings": { "title": "Extract Headings", "type": "boolean", "description": "Extract all h1-h6 headings with their level.", "default": true }, "extractTables": { "title": "Extract Tables", "type": "boolean", "description": "Extract HTML tables as structured data (headers + rows).", "default": true }, "extractStructuredData": { "title": "Extract JSON-LD", "type": "boolean", "description": "Extract JSON-LD structured data (schema.org).", "default": true }, "extractEmails": { "title": "Extract Emails", "type": "boolean", "description": "Find email addresses in the page HTML.", "default": true }, "extractPhones": { "title": "Extract Phone Numbers", "type": "boolean", "description": "Find phone numbers in the page HTML.", "default": true }, "cssSelector": { "title": "CSS Selector", "type": "string", "description": "Target specific content. Returns matching text in selectedContent field. Example: .product-price, #main-content, article" }, "maxDepth": { "title": "Crawl Depth", "minimum": 0, "maximum": 5, "type": "integer", "description": "0 = scrape listed URLs only. 1+ = follow internal links to that depth.", "default": 0 }, "maxPages": { "title": "Max Pages", "minimum": 1, "maximum": 10000, "type": "integer", "description": "Maximum total pages to scrape (including followed links).", "default": 100 }, "usePlaywright": { "title": "Enable JS Rendering (Playwright)", "type": "boolean", "description": "Render JavaScript with a real browser. Required for SPAs (React, Vue, Angular), dynamic content, and sites that load data via JS. Uses more compute than default Cheerio mode.", "default": false }, "waitForSelector": { "title": "Wait for Selector", "type": "string", "description": "CSS selector to wait for before extracting. Only works with Playwright enabled. Example: .product-list, #content-loaded" }, "waitForMs": { "title": "Wait (ms)", "minimum": 0, "maximum": 30000, "type": "integer", "description": "Milliseconds to wait after page load before extracting. Only works with Playwright enabled. Useful for lazy-loading content.", "default": 0 }, "extractMarkdown": { "title": "Extract Markdown", "type": "boolean", "description": "Convert page content to clean Markdown. Strips navigation, footers, and boilerplate when stripBoilerplate is enabled. Ideal for LLM/RAG pipelines.", "default": false }, "stripBoilerplate": { "title": "Strip Boilerplate", "type": "boolean", "description": "Remove nav, header, footer, sidebar, and ads before markdown conversion. Keeps only main content.", "default": true }, "chunkSize": { "title": "Chunk Size (chars)", "minimum": 0, "maximum": 50000, "type": "integer", "description": "Split markdown into chunks of this size. Each chunk includes section heading, index, and content hash. Set to 0 to disable chunking.", "default": 0 }, "chunkOverlap": { "title": "Chunk Overlap (chars)", "minimum": 0, "maximum": 5000, "type": "integer", "description": "Character overlap between consecutive chunks. Prevents losing context at chunk boundaries. Typical: 10-20% of chunk size.", "default": 200 }, "includeGlobs": { "title": "Include URL Patterns", "type": "array", "description": "Only follow links matching these glob patterns when crawling. Example: https://example.com/blog/**", "items": { "type": "string" } }, "excludeGlobs": { "title": "Exclude URL Patterns", "type": "array", "description": "Skip links matching these glob patterns. Example: **/login/**, **/admin/**", "items": { "type": "string" } }, "proxyUrl": { "title": "Proxy URL", "type": "string", "description": "HTTP/SOCKS proxy URL for requests. Format: http://user:pass@host:port or socks5://user:pass@host:port. Use Apify Proxy or your own." }, "searchQuery": { "title": "Search Query", "type": "string", "description": "Search Google and scrape the top results. Leave URLs empty to use search mode. Example: 'best project management tools 2026'" }, "searchLimit": { "title": "Search Results Limit", "minimum": 1, "maximum": 50, "type": "integer", "description": "Maximum number of search results to scrape (default: 10).", "default": 10 }, "actions": { "title": "Browser Actions (Playwright only)", "type": "array", "description": "Actions to perform before extraction. Requires Playwright enabled. JSON array. Types: click (selector), fill (selector + value), scroll (pixels), wait (ms), screenshot. Example: [{\"type\":\"click\",\"selector\":\".load-more\"},{\"type\":\"scroll\",\"pixels\":2000}]" }, "dryRun": { "title": "Dry Run", "type": "boolean", "description": "Scrape pages but don't charge.", "default": false }, "requestRetries": { "title": "Request Retries", "minimum": 0, "maximum": 10, "type": "integer", "description": "Retry transient failures (5xx, network, timeout) this many times before marking the page failed. Permanent failures (404, parse) are not retried.", "default": 3 }, "requestTimeoutSecs": { "title": "Request Timeout (seconds)", "minimum": 5, "maximum": 300, "type": "integer", "description": "Per-request timeout in seconds.", "default": 30 }, "rotateUserAgent": { "title": "Rotate User-Agent", "type": "boolean", "description": "Rotate User-Agent from a real-browser pool (Chrome / Firefox / Safari) per request. Reduces basic anti-bot WAF detection.", "default": true }, "cacheEnabled": { "title": "Cache Results (5min TTL)", "type": "boolean", "description": "Cache page results in-memory keyed by (URL + extract-flags). Same URL within 5 minutes returns cached data with fromCache:true. Useful when the same URLs may appear multiple times in one run (e.g., recursive crawl with shared external pages).", "default": false }, "diagnoseMode": { "title": "Diagnostic Mode (no LLM)", "type": "boolean", "description": "Run a contract-readiness diagnostic before LLM extraction. The actor crawls each URL, scores source readiness, and returns verdicts, blockers, warnings, and a recommended next step. Requires urls[].", "default": false }, "extractContract": { "title": "Extraction Contract", "enum": [ "real-estate-listing" ], "type": "string", "description": "Built-in contract name for diagnostic or extract mode. Current option: real-estate-listing. In extractMode this supplies the schema and validation rules; in diagnoseMode it supplies the readiness signals.", "default": "real-estate-listing" }, "diagnosticReport": { "title": "Save Diagnostic Report", "type": "boolean", "description": "When diagnoseMode=true, also save a buyer-readable Markdown report to the run's key-value store as contract-diagnostic-report.", "default": false }, "extractMode": { "title": "Extract Mode (LLM-powered structured extraction)", "type": "boolean", "description": "Switch from plain crawl to LLM-powered structured extraction. The actor crawls each URL, cleans to markdown, and asks an OpenAI-compatible LLM to return ONLY a JSON object matching your schema/prompt or the selected extraction contract. Output is parsed typed data per URL; contract mode also adds validation.valid and missingRequiredFields. Requires llmBaseUrl + llmModel (or matching Actor environment variables).", "default": false }, "extractJsonSchema": { "title": "Extract: JSON Schema", "type": "object", "description": "JSON Schema describing the desired typed output. The LLM is instructed to return JSON matching this. Either this or extractPrompt is required when extractMode=true. Example: {\"type\":\"object\",\"properties\":{\"productName\":{\"type\":\"string\"},\"price\":{\"type\":\"number\"}},\"required\":[\"productName\"]}" }, "extractPrompt": { "title": "Extract: Natural-language Instruction", "type": "string", "description": "Natural-language extraction instruction (e.g., 'Extract the product name, price, currency, availability'). Used alongside or instead of the JSON schema." }, "llmBaseUrl": { "title": "LLM Endpoint URL", "type": "string", "description": "OpenAI-compatible chat-completions URL. Examples: https://api.openai.com/v1/chat/completions, http://your-llm-host:8080/v1/chat/completions. If left blank, the actor reads THECRAWLER_LLM_BASEURL from the Actor environment variables." }, "llmModel": { "title": "LLM Model", "type": "string", "description": "Model name string the endpoint expects (e.g., gpt-4o-mini, or a .gguf filename for llama.cpp). If blank, reads THECRAWLER_LLM_MODEL env var. The LLM API key (if any) is read from THECRAWLER_LLM_API_KEY Actor env var, never from input — keeps secrets out of run history." }, "llmTemperature": { "title": "LLM Temperature", "type": "string", "description": "Inference temperature as a decimal string ('0', '0.7', '1.0'). 0 is recommended for deterministic extraction. (String editor used because Apify input fields don't natively support floats with custom precision.)", "default": "0" }, "llmMaxTokens": { "title": "LLM Max Output Tokens", "minimum": 100, "maximum": 32000, "type": "integer", "description": "Maximum tokens the LLM may return in the JSON response.", "default": 4000 }, "llmMarkdownCharLimit": { "title": "Markdown Character Limit Sent to LLM", "minimum": 1000, "maximum": 200000, "type": "integer", "description": "How much of the page markdown to send to the LLM. Default 30000 chars (≈7.5k tokens). Bigger = more context but slower and more $$ on hosted APIs.", "default": 30000 } } }, "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 } } } } } } } } } }