**Youtube Search Prompt**

What does YouTube Search Prompt do?

YouTube Search Prompt lets you scrape YouTube videos from a plain-English prompt — no URL building required. Describe what you're looking for (for example, "popular videos about home espresso brewing"), and the Actor uses AI to turn your prompt into multiple targeted YouTube search URLs, scrapes the matching videos, and stores them as a structured dataset.

Under the hood, it chains two Actors together:

1. AI Scraper Input Generator converts your prompt into YouTube search URLs.
2. YouTube Scraper crawls those URLs and extracts the videos with full metadata — titles, view counts, likes, channel info, descriptions, and more.

Because it runs on the Apify platform, you get API access, scheduling, monitoring, and integrations (Make, Zapier, webhooks, Google Drive, and more) out of the box.

Why use YouTube Search Prompt?

• No manual URL construction — describe your intent in natural language, and the AI handles YouTube's search URL format and query variations.
• Better coverage — one prompt becomes several targeted search URL variations, surfacing videos a single query would miss.
• Trend research made simple — schedule a daily run of "AI coding assistant reviews" and watch how coverage of a topic evolves.
• Channel and competitor analysis — compare view counts, subscribers, and engagement across every channel covering your niche.
• Automation-ready — predictable JSON output that plugs straight into data pipelines, dashboards, and NLP workflows — without YouTube Data API quotas.

How to use YouTube Search Prompt

1. Open the Actor in Apify Console.
2. In the Input tab, enter your search prompt in plain English.
3. Optionally adjust Max items to limit how many videos are scraped per generated URL.
4. Set the Actor's memory to 128 MB or less in the run options (the Actor enforces this limit, since the heavy lifting happens in the chained Actors).
5. Click Start and wait for the run to finish.
6. Open the Output tab or the dataset API URL to view and export your results.

Input

Set the input in the Actor's Input tab. Only the prompt is required.

{
    "prompt": "popular videos about home espresso brewing",
    "maxItems": 50
}

Field	Type	Required	Description
prompt	string	Yes	Plain-English description of what to find on YouTube. Topics, products, and questions all work — be specific for best results.
maxItems	integer	No	Maximum number of videos to scrape per generated URL (default 10, maximum 500).

Note: maxItems applies per generated URL, so with the default numberOfUrls of 3 a run can return up to 3× maxItems videos.

Output

Each scraped YouTube video is stored as one item in the default dataset. You can download the dataset in various formats such as JSON, HTML, CSV, or Excel.

Example dataset item (shortened):

{
    "title": "Stromae - Santé (Live From The Tonight Show Starring Jimmy Fallon)",
    "id": "CW7gfrTlr0Y",
    "url": "https://www.youtube.com/watch?v=CW7gfrTlr0Y",
    "thumbnailUrl": "https://i.ytimg.com/vi/CW7gfrTlr0Y/maxresdefault.jpg",
    "viewCount": 35582192,
    "date": "2021-12-21",
    "likes": 512238,
    "channelName": "StromaeVEVO",
    "channelUrl": "http://www.youtube.com/@StromaeVEVO",
    "numberOfSubscribers": 6930000,
    "duration": "00:03:17",
    "commentsCount": 14,
    "text": "Stromae - Santé (Live From The Tonight Show Starring Jimmy Fallon on NBC)...",
    "subtitles": ["en", "fr"],
    "isMonetized": true
}

Data table

Field	Type	Description
title	string	Video title
id	string	YouTube video ID
url	string	Direct link to the video
thumbnailUrl	string	Video thumbnail image URL
viewCount	number	Total view count
date	string	Upload date
likes	number	Like count
channelName / channelUrl	string	Channel name and URL
numberOfSubscribers	number	Channel subscriber count
duration	string	Video length (HH:MM:SS)
text	string	Video description
subtitles	array	Available caption language codes
isMonetized	boolean	Whether the video shows ads

Exact fields may vary slightly by video, since not all YouTube videos expose the same attributes.

How much does it cost to scrape YouTube with a prompt?

A run consists of three parts: this orchestrator Actor (very light — it runs at 128 MB and mostly waits), one short AI Scraper Input Generator run, and the YouTube scraper run whose cost scales with the number of videos collected. For small runs (tens of videos) the total cost is typically well under a dollar. Use Max items and Number of URLs to cap the scraping cost of each run. If you are on the Apify free tier, start with small values to see actual consumption before scaling up.

Tips

• Be specific in your prompt. "Beginner espresso machine reviews under $500" produces much better search URLs than "coffee".
• Start small. Run with a low maxItems first to check that the generated URLs match your intent, then scale up (maximum 500 per URL).
• Schedule it. Add the Actor to a Schedule to monitor a topic or niche daily, and connect a webhook or integration to act on the results.
• Keep memory at 128 MB. The Actor intentionally refuses to run with more, because the chained Actors do the actual work — extra memory would only add cost.

FAQ, disclaimers, and support

Is it legal to scrape YouTube? This Actor extracts publicly available video data. You are responsible for reviewing and complying with YouTube's Terms of Service and applicable laws (including data protection regulations) before using the results, especially in automated or commercial workflows. Do not collect or misuse personal data.

Why did my run fail with a memory error? This Actor requires memory of 128 MB or less. Lower the memory setting in the run options and start it again.

Why are there no results? If the AI could not generate valid search URLs from your prompt, the run fails with a clear error — rephrase the prompt with more specific topic terms and try again.

Found a bug or need a custom feature? Open an issue in the Actor's Issues tab — custom solutions and tailored versions are available on request.