# YouTube Video Reverse Engineer AI Script, Frame & Hook Analysis (`brilliant_gum/youtube-video-reverse-engineer`) Actor Turn any YouTube video into a blueprint. AI extracts hook formulas, script structure, retention techniques, style DNA, and audience engagement from top comments. Outputs are ready-to-use prompts — feed to ChatGPT for scripts or Midjurney for visuals. Premium adds frame-by-frame visual analysis. - **URL**: https://apify.com/brilliant\_gum/youtube-video-reverse-engineer.md - **Developed by:** [Yuliia Kulakova](https://apify.com/brilliant_gum) (community) - **Categories:** AI, Videos - **Stats:** 2 total users, 1 monthly users, 100.0% runs succeeded, NaN bookmarks - **User rating**: No ratings yet ## Pricing from $0.30 / video minute (full) This Actor is paid per event and usage. You are charged both the fixed price for specific events and for Apify platform usage. Learn more: https://docs.apify.com/platform/actors/running/actors-in-store#pay-per-event ## What's an Apify Actor? Actors are a software tools running on the Apify platform, for all kinds of web data extraction and automation use cases. In Batch mode, an Actor accepts a well-defined JSON input, performs an action which can take anything from a few seconds to a few hours, and optionally produces a well-defined JSON output, datasets with results, or files in key-value store. In Standby mode, an Actor provides a web server which can be used as a website, API, or an MCP server. Actors are written with capital "A". ## How to integrate an Actor? If asked about integration, you help developers integrate Actors into their projects. You adapt to their stack and deliver integrations that are safe, well-documented, and production-ready. The best way to integrate Actors is as follows. In JavaScript/TypeScript projects, use official [JavaScript/TypeScript client](https://docs.apify.com/api/client/js.md): ```bash npm install apify-client ``` In Python projects, use official [Python client library](https://docs.apify.com/api/client/python.md): ```bash pip install apify-client ``` In shell scripts, use [Apify CLI](https://docs.apify.com/cli/docs.md): ````bash # MacOS / Linux curl -fsSL https://apify.com/install-cli.sh | bash # Windows irm https://apify.com/install-cli.ps1 | iex ```bash In AI frameworks, you might use the [Apify MCP server](https://docs.apify.com/platform/integrations/mcp.md). If your project is in a different language, use the [REST API](https://docs.apify.com/api/v2.md). For usage examples, see the [API](#api) section below. For more details, see Apify documentation as [Markdown index](https://docs.apify.com/llms.txt) and [Markdown full-text](https://docs.apify.com/llms-full.txt). # README ## YouTube Video Reverse Engineer — AI Script & Hook Analysis ![YouTube Video Reverse Engineer](https://i.imgur.com/txFQJSO.png) Reverse-engineer any YouTube video with AI. Extract the complete content blueprint: script structure, hook formulas, retention techniques, style DNA, pacing analysis, thumbnail psychology, visual production patterns, and ready-to-use recreation prompts. Decode why videos go viral and replicate the exact formula for your content. ### What does this YouTube video analyzer do? This AI-powered YouTube video analysis tool deconstructs videos into their structural DNA. Unlike simple transcript scrapers or metadata extractors, it identifies **why** a video works by mapping every storytelling technique, retention hook, emotional trigger, and visual pattern used throughout the video. Feed it any YouTube URL — from MrBeast challenges to tech tutorials to educational content — and get back a complete structural blueprint you can adapt to your own niche. **Key outputs:** - **Script structure breakdown** — Hook type/timing/score, intro thesis, body segments with content types, CTA analysis, section timing percentages, words-per-minute pacing - **Style DNA extraction** — Niche classification, target audience profiling, sentence rhythm patterns, tone mapping, script flow analysis, transition styles, curiosity gap mechanics with timestamps - **Retention technique identification** — Open loops, pattern interrupts, callbacks, running bits, escalation patterns, emotional beats — each with exact timestamps and mechanical explanations - **Recreation prompts** — A complete AI-ready prompt that captures the video's structural blueprint, plus title formulas, hook templates, and section-by-section instructions - **Visual analysis** (Premium) — Frame-by-frame breakdown: shot types, on-screen text, camera work, lighting, color palette, transitions, visual retention techniques - **Visual Style Profile** (Premium) — Complete visual DNA including color grading, composition patterns, and a master image style prompt for consistent visual recreation - **Thumbnail analysis** (Premium) — Color psychology, composition techniques, emotion detection, curiosity elements, CTR score, plus alternative thumbnail concepts with image prompts ### How it works ```` Step 1: Paste YouTube URL(s) → choose Full or Premium Step 2: AI extracts transcript, analyzes structure, identifies techniques Step 3: Get a complete video blueprint with recreation prompts in structured JSON ```` The actor extracts the transcript (auto-captions or manual), preprocesses it (WPM calculation, pause detection, natural break identification), then runs multi-pass AI analysis to identify structure, map retention techniques, extract style DNA, and generate actionable recreation prompts. Premium tier additionally extracts video frames and runs visual AI analysis to build a complete Visual Style Profile and thumbnail breakdown. ### Pricing | Tier | Base Price | What You Get | |------|-----------|--------------| | **Full** | $0.25 / minute of video | Structure + Style DNA + retention techniques + recreation prompts | | **Premium** | $0.45 / minute of video | Full + thumbnail analysis + frame-by-frame visual analysis + Visual Style Profile | **Frame density multiplier (Premium only):** The default screenshot interval is 15 seconds. Denser intervals (more frames = more AI analysis) cost proportionally more: - 15s interval = base rate ($0.45/min) - 10s interval = +50% on frame analysis surcharge - 5s interval = +200% on frame analysis surcharge **Examples:** - 10-min video, Full depth = **$2.50** - 14-min video, Premium (15s interval) = **$6.30** - 14-min video, Premium (5s interval) = **$11.90** - 45-min video, Full depth = **$11.25** Use **Dry Run** mode to preview the exact cost before processing — returns video metadata, estimated frames, and price with zero charges. ### Input example ```json { "urls": ["https://www.youtube.com/watch?v=fMfipiV_17o"], "language": "en", "analysisDepth": "full", "recreationTopic": "productivity tips for developers", "dryRun": false } ```` | Parameter | Description | Default | |-----------|-------------|---------| | `urls` | YouTube video URLs (batch supported) | required | | `language` | Preferred transcript language (ISO code) | `"en"` | | `analysisDepth` | `"full"` or `"premium"` | `"full"` | | `screenshotIntervalSec` | Premium: frame extraction interval | `15` | | `recreationTopic` | Adapt recreation prompt to your niche | optional | | `dryRun` | Preview cost without processing | `false` | ### Output example (real MrBeast analysis) Analyzed: **"Would You Sit In Snakes For $10,000?"** by MrBeast (14 min, 3,436 words) ```json { "url": "https://www.youtube.com/watch?v=fMfipiV_17o", "videoId": "fMfipiV_17o", "videoTitle": "Would You Sit In Snakes For $10,000?", "channel": "MrBeast", "durationSeconds": 846, "transcriptWordCount": 3436, "avgWordsPerMinute": 244, "analysisDepth": "full", "structure": { "hook": { "type": "immediate_challenge", "start_sec": 0, "end_sec": 18, "summary": "Opens with a bathtub full of snakes, immediate challenge proposition ('sit in this tub for $10K for your mom'), and Chandler instantly leaving before MrBeast finishes. Zero preamble — viewer drops into stakes + visual spectacle + comedy.", "score": 9, "score_reasoning": "Perfect MrBeast hook formula: (1) visual spectacle immediately, (2) money stakes in first 5 seconds, (3) comedy payoff within 10 seconds, (4) emotional stakes (for your mom). Self-contained micro-story optimized for retention." }, "intro": { "start_sec": 18, "end_sec": 35, "thesis": "A compilation of escalating physical/psychological challenges where contestants face fears for cash prizes — testing how far people will go for money." }, "body_segments": [ { "title": "Snake Bathtub ($10K)", "start_sec": 0, "end_sec": 80, "key_points": ["Chandler's instant refusal = comedy gold", "20 snakes added progressively builds tension", "Prize goes to mom (emotional stakes)"] }, { "title": "Cockroach Money Grab ($9,340)", "start_sec": 80, "end_sec": 140, "key_points": ["Random crew member inclusion", "MrBeast's chaotic time-keeping = comedy", "Oddball dollar amount adds authenticity"] } ], "total_body_segments": 11, "overall_structure": "rapid_compilation" }, "style_dna": { "niche": "entertainment / challenge / philanthropy", "target_audience": "Ages 8-25, primarily male, global. YouTube-native viewers who enjoy spectacle, money, and fast-paced entertainment.", "hook_style": "cold_open_spectacle — zero preamble. Visual spectacle + money stakes + comedy all within 15 seconds.", "script_flow": "rapid_compilation — 11+ discrete challenges in 14 minutes (~75 seconds per segment). Each is a self-contained mini-story.", "sentence_rhythm": "staccato_chaotic — rapid-fire dialogue, short interrupted sentences. Universal comprehension across ages and languages.", "tone": "chaotic_generous — oscillates between genuine excitement at giving money, deliberate trolling, and chaotic energy.", "retention_techniques": [ "cold_open — no intro, no logo, first frame is the challenge itself", "running_bit_thread — Noah's curling marathon spans entire video, provides continuity", "segment_variety — no two segments use the same mechanic", "escalating_money — prizes climb: $9K → $10K → $20K → $52K → car", "character_callbacks — Chandler's phobia, Karl's island loss, Chris's competitive streak" ], "curiosity_gaps": [ {"timestamp_sec": 0, "text": "If any of you sits in this tub of snakes...", "mechanic": "immediate_stakes"}, {"timestamp_sec": 400, "text": "I'm gonna count your reps... but I'm not gonna tell you when to stop", "mechanic": "unknown_endpoint"} ] }, "recreation": { "recreation_prompt": "Create a 14-minute rapid-fire challenge compilation video script with 10-12 discrete segments. Follow this structural blueprint:\n\nCOLD OPEN (2%): Open on the most visually spectacular challenge. No intro, no greeting. First sentence states challenge and money stakes. Second beat: someone reacts comically. Third beat: someone does it.\n\nSEGMENTS (96%): 10-12 discrete challenges rotating between types: FEAR, SKILL/ENDURANCE, GUESSING/LUCK, DOUBLE-OR-NOTHING, WHOLESOME SURPRISE...", "title_formulas": [ "Would You [Scary Action] For $[Amount]?", "I Gave [Person] $[Amount] If They [Challenge]", "$[Amount] vs [Scary Thing] — Who Wins?" ], "hook_template": "Open cold on [VISUAL SPECTACLE]. State '$[AMOUNT] if you [CHALLENGE]' within 3 seconds. [PERSON] immediately refuses/fails for comedy. [ANOTHER PERSON] steps up — tension begins." }, "transcript": "[00:00] This is a bathtub full of snakes. Hey there, little guy...\n[00:05] If any of you sits in this tub of snakes, I'll give your mom $10,000...\n..." } ``` *Note: Output truncated for display. Full output includes all 11 body segments, 9 retention techniques, complete recreation prompt with per-section instructions, and full timestamped transcript.* ### Premium tier additional output Premium adds three powerful visual analysis layers on top of Full. Real example from a TED Talk analysis: ```json { "thumbnail_analysis": { "dominant_colors": [ "#8B2F8B (deep magenta — TED curtains)", "#4A4A4A (charcoal — speaker's blazer)", "#6B8FBF (steel blue — shirt)", "#D4A76A (warm skin tones)" ], "text_overlay": "No text overlay on the thumbnail itself.", "faces": { "count": 1, "primary_face": "Middle-aged man with gray curly hair, captured mid-gesture — mouth slightly open, eyes focused. Confident and animated.", "expression": "animated_engaged", "eye_contact": false }, "composition": "Rule of thirds with speaker right-center. Raised hand creates diagonal line. Purple TED curtain provides rich backdrop.", "click_score": 6, "click_reasoning": "Standard TED thumbnail — professional but not optimized for CTR. No text overlay, no eye contact, muted colors.", "thumbnail_concepts": [ { "concept": "Brain Under Siege", "description": "Split-face: left half calm, right half stressed with red tint and chaotic symbols. Text: 'YOUR BRAIN IS LYING TO YOU'", "image_prompt": "Close-up split portrait divided down center. Left: calm, blue-toned, organized. Right: stressed, reddened skin, floating chaotic symbols (clock, pills, keys). Bold white text at bottom. Black background." }, { "concept": "Pre-Mortem Checklist", "description": "Overhead shot of a neat checklist with red X marks through disaster scenarios", "image_prompt": "Top-down flatlay: white paper checklist with handwritten items, several crossed out in red marker. Coffee cup corner. Warm lighting. Text overlay: 'THE CHECKLIST THAT SAVES LIVES'" } ] }, "visual_analysis": [ { "frame_index": 0, "timestamp_sec": 0, "shot_type": "medium_shot — chest-up framing capturing speaker's upper body and hands", "on_screen_text": "None — clean frame", "visual_elements": [ "Speaker center-right in charcoal blazer and steel blue shirt", "Right hand raised near temple (brain/thinking gesture)", "Deep magenta-purple TED curtain filling background", "Metal staircase visible upper-left adding depth" ], "camera_angle": "eye-level, slightly left of center", "camera_movement": "static — locked-off tripod", "lighting": "Professional three-point stage lighting with warm golden skin tones against cool purple backdrop", "color_palette": ["#8B2F8B", "#4A4A4A", "#D4A76A"] } ], "visual_style_profile": { "art_style": "Classic TED Talk production — single-speaker stage with professional multi-camera capture. Clean, polished broadcast quality.", "color_palette": "Deep magenta-purple (55%), charcoal gray (15%), near-black (15%), steel blue (8%), warm gold skin tones (7%).", "lighting_style": "Professional three-point: key light above-front, fill from sides, purple color wash on backdrop.", "camera_work": "Medium shot dominant. Slow, deliberate movements. TED's 'invisible production' philosophy.", "composition_patterns": "Consistent rule-of-thirds. Speaker's hand gesture creates diagonal leading line. Strong negative space balance.", "visual_retention_techniques": "Minimal — content carries attention. Hand gestures provide visual interest at key points.", "master_image_style_prompt": "Professional TED-style stage photography: single speaker on minimalist stage with deep magenta-purple curtain backdrop, warm golden three-point lighting highlighting animated hand gestures. Business-casual wardrobe. Medium shot chest-up framing. Clean, authoritative, educational tone." } } ``` The `master_image_style_prompt` can be fed directly into image generation tools (Midjourney, DALL-E, Stable Diffusion) to recreate the visual style for your own content. ### Who is this for? - **YouTube creators** — Analyze why competitor videos outperform yours. Get the exact hook formula, pacing, and retention techniques in seconds instead of hours of manual note-taking. - **Content agencies** — Deconstruct client competitors at scale. Batch-analyze 50 videos to find recurring patterns that drive views and engagement. - **Scriptwriters** — Extract proven video structure templates. Get hook formulas, emotional arc patterns, and pacing rhythms from any successful video. - **Course creators** — Study what makes educational content engaging. Identify optimal pacing, curiosity gaps, and knowledge delivery patterns. - **SEO and growth teams** — Extract title formulas and hook patterns for data-driven A/B testing of YouTube content strategy. - **AI content creators** — Video-to-prompt pipeline. Get ready-to-use prompts for script generation, image generation, and thumbnail creation. ### Dry run (cost preview) ```json { "urls": ["https://www.youtube.com/watch?v=fMfipiV_17o"], "dryRun": true, "analysisDepth": "premium" } ``` Returns: ```json { "videoId": "fMfipiV_17o", "title": "Would You Sit In Snakes For $10,000?", "channel": "MrBeast", "durationSeconds": 846, "durationMinutes": 15, "transcriptTokens": 5571, "estimatedCost": 6.75, "pricePerMinute": 0.45, "captionsAvailable": true, "dryRun": true } ``` ### Frequently asked questions **How is this different from YouTube transcript scrapers?** Transcript scrapers give you raw text. This actor analyzes the text with AI to extract the structural blueprint — hook formulas, retention techniques, emotional arcs, pacing patterns — and generates actionable recreation prompts. It's the difference between reading sheet music and understanding music theory. **What languages are supported?** Any language with YouTube captions available. Set the `language` parameter to the ISO code (e.g., "es", "de", "ja", "ru", "ko"). Falls back to the first available caption track if your preferred language isn't found. **Can I batch-analyze an entire YouTube channel?** Yes. Pass multiple URLs in the `urls` array to batch-analyze videos. Analyze 10-50 videos from one channel to identify their recurring patterns, hook formulas, and content evolution over time. **What if a video has no captions?** The actor returns a clear error for that video and continues processing the rest of the batch. Most YouTube videos have auto-generated captions available. **How accurate is the AI analysis?** Structure detection aligns within 5-10 seconds of actual transitions. Retention technique identification catches 80-90% of intentional hooks. Recreation prompts produce scripts that match the original's pacing, style, and engagement patterns. **Can I use the recreation prompts with ChatGPT, Claude, or other AI?** Yes. The recreation prompt is designed to be pasted directly into any AI assistant to generate a new script matching the original's structural formula while adapting to your topic. **How long does analysis take?** Full depth: 1-3 minutes per video. Premium depth: 3-7 minutes per video depending on length. The actor handles videos up to 60 minutes. **What's the frame-by-frame visual analysis?** Premium tier extracts screenshots at regular intervals and analyzes each frame for shot composition, camera work, lighting, text overlays, visual effects, and transitions. These are synthesized into a Visual Style Profile for consistent visual recreation. ### Integration and API ```javascript import { ApifyClient } from 'apify-client'; const client = new ApifyClient({ token: 'YOUR_TOKEN' }); const run = await client.actor('brilliant_gum/youtube-video-reverse-engineer').call({ urls: ['https://youtube.com/watch?v=fMfipiV_17o'], analysisDepth: 'full', recreationTopic: 'tech product reviews', }); const { items } = await client.dataset(run.defaultDatasetId).listItems(); console.log(items[0].recreation.recreation_prompt); // → "Create a 14-minute rapid-fire challenge compilation..." console.log(items[0].recreation.title_formulas); // → ["Would You [Action] For $[Amount]?", ...] ``` Works with Apify webhooks — send results to Zapier, Make, n8n, or your own endpoint when analysis completes. ### Related tools - [YouTube Transcript Scraper](https://apify.com/store) — If you only need raw transcripts without analysis - [YouTube Channel Analyzer](https://apify.com/store) — For channel-level metrics and growth analytics # Actor input Schema ## `urls` (type: `array`): One or more YouTube video URLs to analyze. Supports youtube.com/watch?v=, youtu.be/, and embed URLs. ## `language` (type: `string`): Preferred transcript language (ISO 639-1 code). Falls back to first available track if not found. ## `analysisDepth` (type: `string`): Full ($0.25/min): script structure + style DNA + retention techniques + recreation prompts. Premium ($0.45/min): Full + thumbnail analysis + frame-by-frame visual analysis + visual style profile. ## `screenshotIntervalSec` (type: `integer`): Premium only. Extract a video frame every N seconds for visual analysis. Lower = more frames analyzed, more detail, longer processing. ## `recreationTopic` (type: `string`): Adapt the recreation prompt to a specific topic or niche. Example: 'productivity tips for developers'. Leave empty for a generic structural blueprint. ## `dryRun` (type: `boolean`): Preview mode: returns video metadata and estimated cost without running analysis. Zero charges. ## Actor input object example ```json { "urls": [ "https://www.youtube.com/watch?v=fMfipiV_17o" ], "language": "en", "analysisDepth": "full", "screenshotIntervalSec": 15, "dryRun": false } ``` # Actor output Schema ## `videoId` (type: `string`): No description ## `videoTitle` (type: `string`): No description ## `channel` (type: `string`): No description ## `durationSeconds` (type: `string`): No description ## `analysisDepth` (type: `string`): No description ## `framesAnalyzed` (type: `string`): No description # API You can run this Actor programmatically using our API. Below are code examples in JavaScript, Python, and CLI, as well as the OpenAPI specification and MCP server setup. ## JavaScript example ```javascript import { ApifyClient } from 'apify-client'; // Initialize the ApifyClient with your Apify API token // Replace the '' with your token const client = new ApifyClient({ token: '', }); // Prepare Actor input const input = { "urls": [ "https://www.youtube.com/watch?v=fMfipiV_17o" ] }; // Run the Actor and wait for it to finish const run = await client.actor("brilliant_gum/youtube-video-reverse-engineer").call(input); // Fetch and print Actor results from the run's dataset (if any) console.log('Results from dataset'); console.log(`💾 Check your data here: https://console.apify.com/storage/datasets/${run.defaultDatasetId}`); const { items } = await client.dataset(run.defaultDatasetId).listItems(); items.forEach((item) => { console.dir(item); }); // 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/js/docs ``` ## Python example ```python from apify_client import ApifyClient # Initialize the ApifyClient with your Apify API token # Replace '' with your token. client = ApifyClient("") # Prepare the Actor input run_input = { "urls": ["https://www.youtube.com/watch?v=fMfipiV_17o"] } # Run the Actor and wait for it to finish run = client.actor("brilliant_gum/youtube-video-reverse-engineer").call(run_input=run_input) # Fetch and print Actor results from the run's dataset (if there are any) print("💾 Check your data here: https://console.apify.com/storage/datasets/" + run["defaultDatasetId"]) for item in client.dataset(run["defaultDatasetId"]).iterate_items(): print(item) # 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/python/docs/quick-start ``` ## CLI example ```bash echo '{ "urls": [ "https://www.youtube.com/watch?v=fMfipiV_17o" ] }' | apify call brilliant_gum/youtube-video-reverse-engineer --silent --output-dataset ``` ## MCP server setup ```json { "mcpServers": { "apify": { "command": "npx", "args": [ "mcp-remote", "https://mcp.apify.com/?tools=brilliant_gum/youtube-video-reverse-engineer", "--header", "Authorization: Bearer " ] } } } ``` ## OpenAPI specification ```json { "openapi": "3.0.1", "info": { "title": "YouTube Video Reverse Engineer AI Script, Frame & Hook Analysis", "description": "Turn any YouTube video into a blueprint. AI extracts hook formulas, script structure, retention techniques, style DNA, and audience engagement from top comments. Outputs are ready-to-use prompts — feed to ChatGPT for scripts or Midjurney for visuals. Premium adds frame-by-frame visual analysis.", "version": "1.0", "x-build-id": "4KNTzCbIHYJbfxnS1" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/brilliant_gum~youtube-video-reverse-engineer/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-brilliant_gum-youtube-video-reverse-engineer", "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": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/brilliant_gum~youtube-video-reverse-engineer/runs": { "post": { "operationId": "runs-sync-brilliant_gum-youtube-video-reverse-engineer", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/brilliant_gum~youtube-video-reverse-engineer/run-sync": { "post": { "operationId": "run-sync-brilliant_gum-youtube-video-reverse-engineer", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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": "YouTube Video URLs", "minItems": 1, "type": "array", "description": "One or more YouTube video URLs to analyze. Supports youtube.com/watch?v=, youtu.be/, and embed URLs.", "items": { "type": "string" } }, "language": { "title": "Transcript Language", "type": "string", "description": "Preferred transcript language (ISO 639-1 code). Falls back to first available track if not found.", "default": "en" }, "analysisDepth": { "title": "Analysis Depth", "enum": [ "full", "premium" ], "type": "string", "description": "Full ($0.25/min): script structure + style DNA + retention techniques + recreation prompts. Premium ($0.45/min): Full + thumbnail analysis + frame-by-frame visual analysis + visual style profile.", "default": "full" }, "screenshotIntervalSec": { "title": "Screenshot Interval (seconds)", "minimum": 5, "maximum": 120, "type": "integer", "description": "Premium only. Extract a video frame every N seconds for visual analysis. Lower = more frames analyzed, more detail, longer processing.", "default": 15 }, "recreationTopic": { "title": "Recreation Topic (optional)", "type": "string", "description": "Adapt the recreation prompt to a specific topic or niche. Example: 'productivity tips for developers'. Leave empty for a generic structural blueprint." }, "dryRun": { "title": "Dry Run (cost preview)", "type": "boolean", "description": "Preview mode: returns video metadata and estimated cost without running analysis. Zero charges.", "default": false } } }, "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 } } } } } } } } } } ```