Travis Vasceannie
37d1e434af
Based on live API testing of v2 endpoints: ✅ Validated against actual responses: - /v2/scrape - matches schema ✓ - /v2/map - corrected response structure (direct links array with url objects) - /v2/crawl - added missing 'url' field to response - /v2/batch/scrape - added 'url' and 'invalidURLs' fields - /v2/search - corrected to data.web structure with creditsUsed at root 🔧 Schema corrections: - MapResponse: Fixed to use direct 'links' array with {url: string} objects - CrawlResponse: Added required 'url' field for status endpoint - BatchScrapeResponse: Added 'url' and 'invalidURLs' fields - SearchResponse: Fixed to use data.web structure + root creditsUsed - ScrapeMetadata: Added optional title, favicon, language fields 🚮 Removed all v1 endpoints to focus only on v2 API 🏷️ Updated title to 'Firecrawl API v2' and version to '2.0.0' All schemas now accurately reflect actual API behavior from container 5d8e66d87762
1256 lines
47 KiB
JSON
1256 lines
47 KiB
JSON
{
|
|
"openapi": "3.1.0",
|
|
"info": {
|
|
"title": "Firecrawl API v2",
|
|
"version": "2.0.0",
|
|
"description": "Comprehensive web scraping and crawling API v2 with support for single page scraping, batch processing, website crawling, and content extraction."
|
|
},
|
|
"servers": [
|
|
{
|
|
"url": "http://crawl.lab:30002",
|
|
"description": "Local Firecrawl instance"
|
|
}
|
|
],
|
|
"security": [
|
|
{
|
|
"BearerAuth": []
|
|
}
|
|
],
|
|
"tags": [
|
|
{
|
|
"name": "scraping",
|
|
"description": "Single page scraping operations"
|
|
},
|
|
{
|
|
"name": "crawling",
|
|
"description": "Website crawling operations"
|
|
},
|
|
{
|
|
"name": "batch",
|
|
"description": "Batch processing operations"
|
|
},
|
|
{
|
|
"name": "extraction",
|
|
"description": "Content extraction operations"
|
|
},
|
|
{
|
|
"name": "search",
|
|
"description": "Search and discovery operations"
|
|
},
|
|
{
|
|
"name": "jobs",
|
|
"description": "Job status and management"
|
|
},
|
|
{
|
|
"name": "team",
|
|
"description": "Team usage and billing"
|
|
},
|
|
{
|
|
"name": "websockets",
|
|
"description": "WebSocket endpoints for real-time updates"
|
|
}
|
|
],
|
|
"components": {
|
|
"securitySchemes": {
|
|
"BearerAuth": {
|
|
"type": "http",
|
|
"scheme": "bearer",
|
|
"bearerFormat": "JWT",
|
|
"description": "JWT token for authentication"
|
|
}
|
|
},
|
|
"schemas": {
|
|
"Error": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean",
|
|
"const": false
|
|
},
|
|
"error": {
|
|
"type": "string",
|
|
"description": "Error message"
|
|
},
|
|
"code": {
|
|
"type": "string",
|
|
"description": "Error code"
|
|
}
|
|
},
|
|
"required": ["success", "error"]
|
|
},
|
|
"ScrapeRequest": {
|
|
"type": "object",
|
|
"properties": {
|
|
"url": {
|
|
"type": "string",
|
|
"format": "uri",
|
|
"description": "URL to scrape"
|
|
},
|
|
"formats": {
|
|
"type": "array",
|
|
"items": {
|
|
"type": "string",
|
|
"enum": ["markdown", "html", "rawHtml", "screenshot", "links", "summary"]
|
|
},
|
|
"description": "Output formats",
|
|
"default": ["markdown"]
|
|
},
|
|
"onlyMainContent": {
|
|
"type": "boolean",
|
|
"description": "Extract only main content",
|
|
"default": true
|
|
},
|
|
"includeTags": {
|
|
"type": "array",
|
|
"items": {"type": "string"},
|
|
"description": "HTML tags to include"
|
|
},
|
|
"excludeTags": {
|
|
"type": "array",
|
|
"items": {"type": "string"},
|
|
"description": "HTML tags to exclude"
|
|
},
|
|
"waitFor": {
|
|
"type": "integer",
|
|
"description": "Wait time in milliseconds",
|
|
"minimum": 0
|
|
},
|
|
"timeout": {
|
|
"type": "integer",
|
|
"description": "Timeout in milliseconds",
|
|
"minimum": 0,
|
|
"maximum": 600000
|
|
}
|
|
},
|
|
"required": ["url"]
|
|
},
|
|
"ScrapeResponse": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean",
|
|
"const": true
|
|
},
|
|
"data": {
|
|
"type": "object",
|
|
"properties": {
|
|
"markdown": {
|
|
"type": "string",
|
|
"description": "Scraped content in markdown format"
|
|
},
|
|
"html": {
|
|
"type": "string",
|
|
"description": "Scraped content in HTML format"
|
|
},
|
|
"rawHtml": {
|
|
"type": "string",
|
|
"description": "Raw HTML content"
|
|
},
|
|
"screenshot": {
|
|
"type": "string",
|
|
"description": "Base64 encoded screenshot"
|
|
},
|
|
"links": {
|
|
"type": "array",
|
|
"items": {"type": "string"},
|
|
"description": "Extracted links"
|
|
},
|
|
"metadata": {
|
|
"$ref": "#/components/schemas/ScrapeMetadata"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"required": ["success", "data"]
|
|
},
|
|
"ScrapeMetadata": {
|
|
"type": "object",
|
|
"properties": {
|
|
"url": {
|
|
"type": "string",
|
|
"format": "uri",
|
|
"description": "Original URL"
|
|
},
|
|
"scrapeId": {
|
|
"type": "string",
|
|
"description": "Unique scrape identifier"
|
|
},
|
|
"sourceURL": {
|
|
"type": "string",
|
|
"format": "uri",
|
|
"description": "Source URL"
|
|
},
|
|
"statusCode": {
|
|
"type": "integer",
|
|
"description": "HTTP status code"
|
|
},
|
|
"contentType": {
|
|
"type": "string",
|
|
"description": "Content type"
|
|
},
|
|
"creditsUsed": {
|
|
"type": "integer",
|
|
"description": "Credits consumed"
|
|
},
|
|
"proxyUsed": {
|
|
"type": "string",
|
|
"description": "Proxy type used"
|
|
},
|
|
"title": {
|
|
"type": "string",
|
|
"description": "Page title"
|
|
},
|
|
"favicon": {
|
|
"type": "string",
|
|
"format": "uri",
|
|
"description": "Favicon URL"
|
|
},
|
|
"language": {
|
|
"type": "string",
|
|
"description": "Page language"
|
|
}
|
|
}
|
|
},
|
|
"CrawlRequest": {
|
|
"type": "object",
|
|
"properties": {
|
|
"url": {
|
|
"type": "string",
|
|
"format": "uri",
|
|
"description": "Starting URL for crawling"
|
|
},
|
|
"maxDepth": {
|
|
"type": "integer",
|
|
"description": "Maximum crawl depth",
|
|
"minimum": 0,
|
|
"maximum": 10
|
|
},
|
|
"limit": {
|
|
"type": "integer",
|
|
"description": "Maximum number of pages to crawl",
|
|
"minimum": 1,
|
|
"maximum": 10000
|
|
},
|
|
"allowExternalLinks": {
|
|
"type": "boolean",
|
|
"description": "Allow crawling external domains",
|
|
"default": false
|
|
},
|
|
"includePaths": {
|
|
"type": "array",
|
|
"items": {"type": "string"},
|
|
"description": "Paths to include"
|
|
},
|
|
"excludePaths": {
|
|
"type": "array",
|
|
"items": {"type": "string"},
|
|
"description": "Paths to exclude"
|
|
},
|
|
"scrapeOptions": {
|
|
"$ref": "#/components/schemas/ScrapeRequest"
|
|
}
|
|
},
|
|
"required": ["url"]
|
|
},
|
|
"CrawlResponse": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean",
|
|
"const": true
|
|
},
|
|
"id": {
|
|
"type": "string",
|
|
"description": "Crawl job ID"
|
|
},
|
|
"url": {
|
|
"type": "string",
|
|
"format": "uri",
|
|
"description": "URL to check job status"
|
|
}
|
|
},
|
|
"required": ["success", "id", "url"]
|
|
},
|
|
"BatchScrapeRequest": {
|
|
"type": "object",
|
|
"properties": {
|
|
"urls": {
|
|
"type": "array",
|
|
"items": {
|
|
"type": "string",
|
|
"format": "uri"
|
|
},
|
|
"description": "URLs to scrape",
|
|
"maxItems": 1000
|
|
},
|
|
"scrapeOptions": {
|
|
"$ref": "#/components/schemas/ScrapeRequest"
|
|
}
|
|
},
|
|
"required": ["urls"]
|
|
},
|
|
"BatchScrapeResponse": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean",
|
|
"const": true
|
|
},
|
|
"id": {
|
|
"type": "string",
|
|
"description": "Batch scrape job ID"
|
|
},
|
|
"url": {
|
|
"type": "string",
|
|
"format": "uri",
|
|
"description": "URL to check job status"
|
|
},
|
|
"invalidURLs": {
|
|
"type": "array",
|
|
"items": {"type": "string"},
|
|
"description": "URLs that were invalid"
|
|
}
|
|
},
|
|
"required": ["success", "id", "url", "invalidURLs"]
|
|
},
|
|
"MapRequest": {
|
|
"type": "object",
|
|
"properties": {
|
|
"url": {
|
|
"type": "string",
|
|
"format": "uri",
|
|
"description": "Website URL to map"
|
|
},
|
|
"search": {
|
|
"type": "string",
|
|
"description": "Filter URLs by search term"
|
|
},
|
|
"limit": {
|
|
"type": "integer",
|
|
"description": "Maximum number of URLs to return",
|
|
"minimum": 1,
|
|
"maximum": 10000,
|
|
"default": 100
|
|
},
|
|
"includeSubdomains": {
|
|
"type": "boolean",
|
|
"description": "Include subdomains in mapping",
|
|
"default": false
|
|
}
|
|
},
|
|
"required": ["url"]
|
|
},
|
|
"MapResponse": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean",
|
|
"const": true
|
|
},
|
|
"links": {
|
|
"type": "array",
|
|
"items": {
|
|
"type": "object",
|
|
"properties": {
|
|
"url": {
|
|
"type": "string",
|
|
"format": "uri"
|
|
}
|
|
},
|
|
"required": ["url"]
|
|
},
|
|
"description": "Discovered URLs"
|
|
}
|
|
},
|
|
"required": ["success", "links"]
|
|
},
|
|
"SearchRequest": {
|
|
"type": "object",
|
|
"properties": {
|
|
"query": {
|
|
"type": "string",
|
|
"description": "Search query"
|
|
},
|
|
"limit": {
|
|
"type": "integer",
|
|
"description": "Maximum number of results",
|
|
"minimum": 1,
|
|
"maximum": 100,
|
|
"default": 10
|
|
},
|
|
"scrapeOptions": {
|
|
"$ref": "#/components/schemas/ScrapeRequest"
|
|
}
|
|
},
|
|
"required": ["query"]
|
|
},
|
|
"SearchResponse": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean",
|
|
"const": true
|
|
},
|
|
"data": {
|
|
"type": "object",
|
|
"properties": {
|
|
"web": {
|
|
"type": "array",
|
|
"items": {
|
|
"type": "object",
|
|
"properties": {
|
|
"url": {
|
|
"type": "string",
|
|
"format": "uri"
|
|
},
|
|
"title": {
|
|
"type": "string"
|
|
},
|
|
"description": {
|
|
"type": "string"
|
|
},
|
|
"markdown": {
|
|
"type": "string"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"required": ["web"]
|
|
},
|
|
"creditsUsed": {
|
|
"type": "integer",
|
|
"description": "Credits consumed"
|
|
}
|
|
},
|
|
"required": ["success", "data", "creditsUsed"]
|
|
},
|
|
"ExtractRequest": {
|
|
"type": "object",
|
|
"properties": {
|
|
"urls": {
|
|
"type": "array",
|
|
"items": {
|
|
"type": "string",
|
|
"format": "uri"
|
|
},
|
|
"description": "URLs to extract from"
|
|
},
|
|
"prompt": {
|
|
"type": "string",
|
|
"description": "Extraction prompt"
|
|
},
|
|
"schema": {
|
|
"type": "object",
|
|
"description": "JSON schema for extraction"
|
|
}
|
|
},
|
|
"required": ["urls"]
|
|
},
|
|
"ExtractResponse": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean",
|
|
"const": true
|
|
},
|
|
"id": {
|
|
"type": "string",
|
|
"description": "Extract job ID"
|
|
}
|
|
},
|
|
"required": ["success", "id"]
|
|
},
|
|
"JobStatus": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean"
|
|
},
|
|
"status": {
|
|
"type": "string",
|
|
"enum": ["waiting", "active", "completed", "failed", "cancelled", "paused"]
|
|
},
|
|
"total": {
|
|
"type": "integer",
|
|
"description": "Total number of pages/items"
|
|
},
|
|
"completed": {
|
|
"type": "integer",
|
|
"description": "Number of completed pages/items"
|
|
},
|
|
"creditsUsed": {
|
|
"type": "integer",
|
|
"description": "Credits consumed"
|
|
},
|
|
"expiresAt": {
|
|
"type": "string",
|
|
"format": "date-time",
|
|
"description": "Job expiration time"
|
|
},
|
|
"data": {
|
|
"type": "array",
|
|
"items": {
|
|
"type": "object",
|
|
"properties": {
|
|
"markdown": {
|
|
"type": "string"
|
|
},
|
|
"html": {
|
|
"type": "string"
|
|
},
|
|
"rawHtml": {
|
|
"type": "string"
|
|
},
|
|
"metadata": {
|
|
"$ref": "#/components/schemas/ScrapeMetadata"
|
|
}
|
|
}
|
|
},
|
|
"description": "Job results (when completed)"
|
|
}
|
|
},
|
|
"required": ["success", "status"]
|
|
},
|
|
"CreditUsage": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean",
|
|
"const": true
|
|
},
|
|
"data": {
|
|
"type": "object",
|
|
"properties": {
|
|
"totalCreditsUsed": {
|
|
"type": "integer"
|
|
},
|
|
"remainingCredits": {
|
|
"type": "integer"
|
|
},
|
|
"totalCreditsPurchased": {
|
|
"type": "integer"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"required": ["success", "data"]
|
|
}
|
|
}
|
|
},
|
|
"paths": {
|
|
"/v2/scrape": {
|
|
"post": {
|
|
"tags": ["scraping"],
|
|
"summary": "Scrape a single URL",
|
|
"description": "Extract content from a single web page in various formats",
|
|
"operationId": "scrapeUrl",
|
|
"requestBody": {
|
|
"required": true,
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/ScrapeRequest"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"responses": {
|
|
"200": {
|
|
"description": "Successful scrape",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/ScrapeResponse"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"400": {
|
|
"description": "Bad request",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/Error"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"401": {
|
|
"description": "Unauthorized",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/Error"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"408": {
|
|
"description": "Request timeout",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/Error"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"500": {
|
|
"description": "Internal server error",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/Error"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/scrape/{jobId}": {
|
|
"get": {
|
|
"tags": ["jobs"],
|
|
"summary": "Get scrape job status",
|
|
"description": "Retrieve the status and results of a scrape job",
|
|
"operationId": "getScrapeStatus",
|
|
"parameters": [
|
|
{
|
|
"name": "jobId",
|
|
"in": "path",
|
|
"required": true,
|
|
"schema": {
|
|
"type": "string"
|
|
},
|
|
"description": "Scrape job ID"
|
|
}
|
|
],
|
|
"responses": {
|
|
"200": {
|
|
"description": "Job status retrieved",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/JobStatus"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"404": {
|
|
"description": "Job not found",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/Error"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/crawl": {
|
|
"post": {
|
|
"tags": ["crawling"],
|
|
"summary": "Start a crawl job",
|
|
"description": "Crawl a website starting from a given URL",
|
|
"operationId": "startCrawl",
|
|
"requestBody": {
|
|
"required": true,
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/CrawlRequest"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"responses": {
|
|
"202": {
|
|
"description": "Crawl job started",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/CrawlResponse"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"400": {
|
|
"description": "Bad request",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/Error"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"401": {
|
|
"description": "Unauthorized",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/Error"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/crawl/{jobId}": {
|
|
"get": {
|
|
"tags": ["jobs"],
|
|
"summary": "Get crawl job status",
|
|
"description": "Retrieve the status and results of a crawl job",
|
|
"operationId": "getCrawlStatus",
|
|
"parameters": [
|
|
{
|
|
"name": "jobId",
|
|
"in": "path",
|
|
"required": true,
|
|
"schema": {
|
|
"type": "string"
|
|
},
|
|
"description": "Crawl job ID"
|
|
}
|
|
],
|
|
"responses": {
|
|
"200": {
|
|
"description": "Job status retrieved",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/JobStatus"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"404": {
|
|
"description": "Job not found",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/Error"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"delete": {
|
|
"tags": ["jobs"],
|
|
"summary": "Cancel crawl job",
|
|
"description": "Cancel a running crawl job",
|
|
"operationId": "cancelCrawl",
|
|
"parameters": [
|
|
{
|
|
"name": "jobId",
|
|
"in": "path",
|
|
"required": true,
|
|
"schema": {
|
|
"type": "string"
|
|
},
|
|
"description": "Crawl job ID"
|
|
}
|
|
],
|
|
"responses": {
|
|
"200": {
|
|
"description": "Job cancelled",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean",
|
|
"const": true
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"404": {
|
|
"description": "Job not found",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/Error"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/crawl/ongoing": {
|
|
"get": {
|
|
"tags": ["jobs"],
|
|
"summary": "Get ongoing crawls",
|
|
"description": "List all ongoing crawl jobs",
|
|
"operationId": "getOngoingCrawls",
|
|
"responses": {
|
|
"200": {
|
|
"description": "List of ongoing crawls",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean"
|
|
},
|
|
"data": {
|
|
"type": "array",
|
|
"items": {
|
|
"$ref": "#/components/schemas/JobStatus"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/crawl/active": {
|
|
"get": {
|
|
"tags": ["jobs"],
|
|
"summary": "Get active crawls",
|
|
"description": "List all active crawl jobs (same as ongoing)",
|
|
"operationId": "getActiveCrawls",
|
|
"responses": {
|
|
"200": {
|
|
"description": "List of active crawls",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean"
|
|
},
|
|
"data": {
|
|
"type": "array",
|
|
"items": {
|
|
"$ref": "#/components/schemas/JobStatus"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/batch/scrape": {
|
|
"post": {
|
|
"tags": ["batch"],
|
|
"summary": "Start batch scrape job",
|
|
"description": "Scrape multiple URLs in batch",
|
|
"operationId": "batchScrape",
|
|
"requestBody": {
|
|
"required": true,
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/BatchScrapeRequest"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"responses": {
|
|
"202": {
|
|
"description": "Batch scrape job started",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/BatchScrapeResponse"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/batch/scrape/{jobId}": {
|
|
"get": {
|
|
"tags": ["jobs"],
|
|
"summary": "Get batch scrape status",
|
|
"description": "Retrieve the status and results of a batch scrape job",
|
|
"operationId": "getBatchScrapeStatus",
|
|
"parameters": [
|
|
{
|
|
"name": "jobId",
|
|
"in": "path",
|
|
"required": true,
|
|
"schema": {
|
|
"type": "string"
|
|
},
|
|
"description": "Batch scrape job ID"
|
|
}
|
|
],
|
|
"responses": {
|
|
"200": {
|
|
"description": "Job status retrieved",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/JobStatus"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"delete": {
|
|
"tags": ["jobs"],
|
|
"summary": "Cancel batch scrape job",
|
|
"description": "Cancel a running batch scrape job",
|
|
"operationId": "cancelBatchScrape",
|
|
"parameters": [
|
|
{
|
|
"name": "jobId",
|
|
"in": "path",
|
|
"required": true,
|
|
"schema": {
|
|
"type": "string"
|
|
},
|
|
"description": "Batch scrape job ID"
|
|
}
|
|
],
|
|
"responses": {
|
|
"200": {
|
|
"description": "Job cancelled",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean",
|
|
"const": true
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/map": {
|
|
"post": {
|
|
"tags": ["search"],
|
|
"summary": "Map website URLs",
|
|
"description": "Discover and map all URLs on a website",
|
|
"operationId": "mapWebsite",
|
|
"requestBody": {
|
|
"required": true,
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/MapRequest"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"responses": {
|
|
"200": {
|
|
"description": "Website mapped successfully",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/MapResponse"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/search": {
|
|
"post": {
|
|
"tags": ["search"],
|
|
"summary": "Search and scrape",
|
|
"description": "Search the web and scrape results",
|
|
"operationId": "search",
|
|
"requestBody": {
|
|
"required": true,
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/SearchRequest"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"responses": {
|
|
"200": {
|
|
"description": "Search results",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/SearchResponse"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/extract": {
|
|
"post": {
|
|
"tags": ["extraction"],
|
|
"summary": "Extract structured data",
|
|
"description": "Extract structured data from web pages using AI",
|
|
"operationId": "extract",
|
|
"requestBody": {
|
|
"required": true,
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/ExtractRequest"
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"responses": {
|
|
"202": {
|
|
"description": "Extract job started",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/ExtractResponse"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/extract/{jobId}": {
|
|
"get": {
|
|
"tags": ["jobs"],
|
|
"summary": "Get extract job status",
|
|
"description": "Retrieve the status and results of an extract job",
|
|
"operationId": "getExtractStatus",
|
|
"parameters": [
|
|
{
|
|
"name": "jobId",
|
|
"in": "path",
|
|
"required": true,
|
|
"schema": {
|
|
"type": "string"
|
|
},
|
|
"description": "Extract job ID"
|
|
}
|
|
],
|
|
"responses": {
|
|
"200": {
|
|
"description": "Job status retrieved",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/JobStatus"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/team/credit-usage": {
|
|
"get": {
|
|
"tags": ["team"],
|
|
"summary": "Get credit usage",
|
|
"description": "Retrieve team credit usage statistics",
|
|
"operationId": "getCreditUsage",
|
|
"responses": {
|
|
"200": {
|
|
"description": "Credit usage data",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/CreditUsage"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/team/credit-usage/historical": {
|
|
"get": {
|
|
"tags": ["team"],
|
|
"summary": "Get historical credit usage",
|
|
"description": "Retrieve historical team credit usage statistics",
|
|
"operationId": "getHistoricalCreditUsage",
|
|
"responses": {
|
|
"200": {
|
|
"description": "Historical credit usage data",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"$ref": "#/components/schemas/CreditUsage"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/team/token-usage": {
|
|
"get": {
|
|
"tags": ["team"],
|
|
"summary": "Get token usage",
|
|
"description": "Retrieve team token usage statistics",
|
|
"operationId": "getTokenUsage",
|
|
"responses": {
|
|
"200": {
|
|
"description": "Token usage data",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean"
|
|
},
|
|
"data": {
|
|
"type": "object",
|
|
"properties": {
|
|
"totalTokensUsed": {
|
|
"type": "integer"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/team/token-usage/historical": {
|
|
"get": {
|
|
"tags": ["team"],
|
|
"summary": "Get historical token usage",
|
|
"description": "Retrieve historical team token usage statistics",
|
|
"operationId": "getHistoricalTokenUsage",
|
|
"responses": {
|
|
"200": {
|
|
"description": "Historical token usage data",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean"
|
|
},
|
|
"data": {
|
|
"type": "object",
|
|
"properties": {
|
|
"totalTokensUsed": {
|
|
"type": "integer"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/concurrency-check": {
|
|
"get": {
|
|
"tags": ["jobs"],
|
|
"summary": "Check concurrency limits",
|
|
"description": "Check current concurrency usage and limits",
|
|
"operationId": "checkConcurrency",
|
|
"responses": {
|
|
"200": {
|
|
"description": "Concurrency status",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean"
|
|
},
|
|
"data": {
|
|
"type": "object",
|
|
"properties": {
|
|
"current": {
|
|
"type": "integer",
|
|
"description": "Current concurrent jobs"
|
|
},
|
|
"limit": {
|
|
"type": "integer",
|
|
"description": "Maximum concurrent jobs allowed"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
},
|
|
"/v2/team/queue-status": {
|
|
"get": {
|
|
"tags": ["team"],
|
|
"summary": "Get queue status",
|
|
"description": "Retrieve current queue status and statistics",
|
|
"operationId": "getQueueStatus",
|
|
"responses": {
|
|
"200": {
|
|
"description": "Queue status data",
|
|
"content": {
|
|
"application/json": {
|
|
"schema": {
|
|
"type": "object",
|
|
"properties": {
|
|
"success": {
|
|
"type": "boolean"
|
|
},
|
|
"data": {
|
|
"type": "object",
|
|
"properties": {
|
|
"waiting": {
|
|
"type": "integer"
|
|
},
|
|
"active": {
|
|
"type": "integer"
|
|
},
|
|
"completed": {
|
|
"type": "integer"
|
|
},
|
|
"failed": {
|
|
"type": "integer"
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|