{ "openapi": "3.0.1", "info": { "title": "YouTube Intelligence Scraper", "description": "Extract public YouTube videos, channels, playlists, Shorts, livestreams, comments, and subtitles into normalized intelligence records", "version": "0.1", "x-build-id": "XeYeOwwn96iq5j5GC" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/qaseemiqbal~youtube-intelligence-scraper/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-qaseemiqbal-youtube-intelligence-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": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/qaseemiqbal~youtube-intelligence-scraper/runs": { "post": { "operationId": "runs-sync-qaseemiqbal-youtube-intelligence-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/qaseemiqbal~youtube-intelligence-scraper/run-sync": { "post": { "operationId": "run-sync-qaseemiqbal-youtube-intelligence-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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", "properties": { "startUrls": { "title": "YouTube URLs", "type": "array", "description": "Video, Shorts, channel, playlist, search result, hashtag, or live URLs.", "items": { "type": "object", "required": [ "url" ], "properties": { "url": { "type": "string", "title": "URL", "description": "Public YouTube URL to scrape." }, "userData": { "type": "object", "title": "User data", "description": "Optional metadata preserved in output source context." } } }, "default": [] }, "searchQueries": { "title": "Search queries", "type": "array", "description": "Search terms exactly as they would be entered in YouTube search.", "items": { "type": "string", "minLength": 1 }, "default": [] }, "channelHandles": { "title": "Channel handles or IDs", "type": "array", "description": "Examples: @Apify, Apify, UCxxxx.", "items": { "type": "string", "minLength": 1 }, "default": [] }, "hashtags": { "title": "Hashtags", "type": "array", "description": "Hashtags to scrape without the # symbol.", "items": { "type": "string", "minLength": 1 }, "default": [] }, "mode": { "title": "Scraping mode", "enum": [ "auto", "search", "videos", "channels", "playlists", "shorts", "live", "comments", "transcripts", "hashtags", "monitor" ], "type": "string", "description": "Choose whether to auto-detect each source type or force a specific scraping workflow.", "default": "auto" }, "maxResultsPerSource": { "title": "Maximum results per source", "minimum": 1, "maximum": 20000, "type": "integer", "description": "Maximum number of main results to collect from each URL, search query, channel, playlist, or hashtag. The default is intentionally small to keep runs cheap.", "default": 25 }, "maxVideos": { "title": "Maximum regular videos", "minimum": 0, "maximum": 20000, "type": "integer", "description": "Maximum number of regular videos to collect per source when video-type limits apply.", "default": 25 }, "maxShorts": { "title": "Maximum Shorts", "minimum": 0, "maximum": 20000, "type": "integer", "description": "Maximum number of YouTube Shorts to collect per source when Shorts are available.", "default": 10 }, "maxStreams": { "title": "Maximum live streams", "minimum": 0, "maximum": 20000, "type": "integer", "description": "Maximum number of live, upcoming, or past livestream videos to collect per source.", "default": 5 }, "maxCommentsPerVideo": { "title": "Maximum comments per video", "minimum": 0, "maximum": 50000, "type": "integer", "description": "Maximum number of top-level public comments to collect for each video when comments are enabled.", "default": 0 }, "maxRepliesPerComment": { "title": "Maximum replies per comment", "minimum": 0, "maximum": 1000, "type": "integer", "description": "Maximum number of public replies to collect for each top-level comment.", "default": 0 }, "includeVideoDetails": { "title": "Include video details", "type": "boolean", "description": "Open each video page to collect richer metadata. Leave disabled for the cheapest discovery runs.", "default": false }, "includeChannelDetails": { "title": "Include channel details", "type": "boolean", "description": "Open channel pages to collect public channel profile information. Leave disabled for lower-cost runs.", "default": false }, "includeDescriptionLinks": { "title": "Extract description links", "type": "boolean", "description": "Extract public links found in video descriptions.", "default": true }, "includeSocialLinks": { "title": "Extract social links", "type": "boolean", "description": "Detect common public social profile links from channel and description links.", "default": true }, "includeComments": { "title": "Include comments", "type": "boolean", "description": "Collect public comments for videos.", "default": false }, "includeReplies": { "title": "Include comment replies", "type": "boolean", "description": "Collect public replies under top-level comments when comments are enabled.", "default": false }, "commentsSort": { "title": "Comments sort", "enum": [ "top", "newest" ], "type": "string", "description": "Preferred public comment ordering when the page supports it.", "default": "top" }, "includeSubtitles": { "title": "Include subtitles/transcripts", "type": "boolean", "description": "Collect public subtitles or transcripts when they are available.", "default": false }, "subtitleLanguages": { "title": "Subtitle languages", "type": "array", "description": "Preferred subtitle language codes, such as en, de, es, or fr.", "items": { "type": "string", "minLength": 2 }, "default": [ "en" ] }, "subtitleFormat": { "title": "Subtitle format", "enum": [ "plainText", "srt", "webvtt", "xml", "json" ], "type": "string", "description": "Format used for saved subtitle or transcript output.", "default": "plainText" }, "includeTranscriptSegments": { "title": "Include transcript segments", "type": "boolean", "description": "Include timestamped transcript segments in transcript records.", "default": false }, "saveSubtitlesToKeyValueStore": { "title": "Save subtitles to Key-Value Store", "type": "boolean", "description": "Save full subtitle files to Apify Key-Value Store and include references in dataset records.", "default": true }, "searchSort": { "title": "Search sort", "enum": [ "relevance", "uploadDate", "viewCount", "rating" ], "type": "string", "description": "Preferred sorting order for YouTube search results.", "default": "relevance" }, "searchDateFilter": { "title": "Search date filter", "enum": [ "any", "hour", "today", "week", "month", "year" ], "type": "string", "description": "Limit YouTube search results by approximate upload date.", "default": "any" }, "searchType": { "title": "Search result type", "enum": [ "all", "video", "channel", "playlist" ], "type": "string", "description": "Restrict YouTube search to videos, channels, playlists, or all supported result types.", "default": "video" }, "durationFilter": { "title": "Duration filter", "enum": [ "any", "short", "medium", "long" ], "type": "string", "description": "Restrict search results by video duration when supported by YouTube.", "default": "any" }, "features": { "title": "Search features", "type": "array", "description": "Optional YouTube search feature filters such as live, HD, 4K, subtitles, or Creative Commons.", "items": { "type": "string", "enum": [ "live", "4k", "hd", "subtitles", "creativeCommons", "360", "vr180", "3d", "hdr", "location", "purchased" ] }, "default": [] }, "channelTab": { "title": "Channel tab", "enum": [ "videos", "shorts", "streams", "playlists", "about", "all" ], "type": "string", "description": "Channel section to scrape, such as videos, Shorts, streams, playlists, about, or all supported tabs.", "default": "videos" }, "channelVideoSort": { "title": "Channel video sort", "enum": [ "newest", "popular", "oldest" ], "type": "string", "description": "Preferred sort order for channel videos when available.", "default": "newest" }, "publishedAfter": { "title": "Published after", "type": "string", "description": "Only include videos published on or after this date in YYYY-MM-DD format." }, "publishedBefore": { "title": "Published before", "type": "string", "description": "Only include videos published on or before this date in YYYY-MM-DD format." }, "language": { "title": "Language", "type": "string", "description": "Preferred YouTube interface and result language.", "default": "en" }, "countryCode": { "title": "Country code", "minLength": 2, "maxLength": 2, "type": "string", "description": "Preferred two-letter country or region code for localized YouTube results.", "default": "US" }, "timezone": { "title": "Timezone", "type": "string", "description": "Timezone used for interpreting dates and run summaries.", "default": "UTC" }, "deduplicate": { "title": "Deduplicate results", "type": "boolean", "description": "Merge duplicate videos found from multiple sources into one canonical result.", "default": true }, "dedupeKey": { "title": "Deduplication key", "enum": [ "videoId", "canonicalUrl", "sourceAndVideoId" ], "type": "string", "description": "Field used to decide whether two video records are duplicates.", "default": "videoId" }, "datasetStrategy": { "title": "Dataset strategy", "enum": [ "single", "separate" ], "type": "string", "description": "Choose whether to store all record types in one dataset or split them into separate datasets.", "default": "single" }, "outputGranularity": { "title": "Output granularity", "enum": [ "flat", "nested", "both" ], "type": "string", "description": "Choose a flatter output for spreadsheets, nested output for APIs, or both.", "default": "nested" }, "errorHandling": { "title": "Error handling", "enum": [ "pushErrorItems", "skipRecoverableErrors", "failOnCritical" ], "type": "string", "description": "Choose how the Actor should handle invalid sources and recoverable page errors.", "default": "pushErrorItems" }, "proxyConfiguration": { "title": "Proxy configuration", "type": "object", "description": "Proxy settings used for YouTube requests. Apify Proxy is recommended.", "default": { "useApifyProxy": true } }, "minConcurrency": { "title": "Minimum concurrency", "minimum": 1, "maximum": 20, "type": "integer", "description": "Minimum number of pages processed in parallel.", "default": 1 }, "maxConcurrency": { "title": "Maximum concurrency", "minimum": 1, "maximum": 20, "type": "integer", "description": "Maximum number of pages processed in parallel. Lower values are usually cheaper and less likely to trigger blocking.", "default": 2 }, "maxRequestRetries": { "title": "Maximum request retries", "minimum": 0, "maximum": 10, "type": "integer", "description": "Maximum number of retries for failed page requests. The default retries once to avoid repeating blocked or very slow requests too many times.", "default": 1 }, "requestTimeoutSecs": { "title": "Request timeout seconds", "minimum": 15, "maximum": 600, "type": "integer", "description": "Maximum time allowed for processing one request, including page navigation and parsing.", "default": 90 }, "navigationTimeoutSecs": { "title": "Navigation timeout seconds", "minimum": 10, "maximum": 180, "type": "integer", "description": "Maximum time allowed for browser navigation. YouTube channel pages can sometimes need more than 30 seconds on cloud proxies.", "default": 45 }, "maxSessionUses": { "title": "Maximum session uses", "minimum": 1, "maximum": 500, "type": "integer", "description": "Retire a browser/proxy session after this many successful requests.", "default": 25 }, "saveSnapshots": { "title": "Save debug snapshots", "type": "boolean", "description": "Save optional HTML snapshots for failed pages to help with debugging.", "default": false }, "debugMode": { "title": "Debug mode", "type": "boolean", "description": "Enable more detailed logs and debug metadata.", "default": false }, "logLevel": { "title": "Log level", "enum": [ "DEBUG", "INFO", "WARNING", "ERROR" ], "type": "string", "description": "Choose how much logging the Actor should produce.", "default": "INFO" } } }, "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 } } } } } } } } } }