YouTube Shorts contains valuable data -- short video results, video ids, view counts, published time, and more. Scraping this data directly means dealing with anti-bot detection, CAPTCHAs, IP rotation, and constantly breaking selectors. The Scavio API handles all of that and returns clean, structured JSON from a single POST request.
This tutorial shows you how to scrape YouTube Shorts using TypeScript and the Scavio API. By the end, you will have a working TypeScript script that fetches real-time YouTube Shorts data and parses the results.
Prerequisites
- TypeScript installed on your machine
- A Scavio API key (free tier includes 50 credits on signup -- no credit card required)
Step 1: Install Dependencies
Install fetch to make HTTP requests:
npm install -D typescript tsxStep 2: Make Your First YouTube Shorts Search
Send a POST request to the Scavio YouTube Shorts API endpoint with your query. The API returns structured JSON with short video results, video ids, view counts, and more.
// Shorts search takes search (not query). channel.name is often empty on Shorts rows -
// resolve it with POST /api/v1/youtube/video when you need it.
const API_KEY = "sk_live_your_key";
const search = "ai coding agent shorts";
interface YouTubeShortsResponse {
data: { results: Array<{ video_id: string; title: string; url: string; duration_text: string | null; view_count: number; published_time: string; channel: { id: string | null; name: string; url: string | null } }>; next_cursor: string | null; has_more: boolean };
response_time: number;
credits_used: number;
credits_remaining: number;
}
const response = await fetch("https://api.scavio.dev/api/v1/youtube/shorts", {
method: "POST",
headers: {
"Authorization": "Bearer " + API_KEY,
"Content-Type": "application/json",
},
body: JSON.stringify({ search }),
});
if (!response.ok) {
throw new Error("Scavio API error: " + response.status);
}
const data = (await response.json()) as YouTubeShortsResponse;
const rows = data?.data?.results ?? [];
for (const row of rows.slice(0, 5)) {
console.log(row?.title);
console.log(" ", row?.view_count, row?.url, row?.video_id);
}Step 3: Example Response
The API returns structured JSON. Here is an example response for a YouTube Shorts search:
{
"data": {
"results": [
{
"video_id": "_P63NrBN49U",
"title": "My go to Full Body Workout. No equipment. Anywhere",
"url": "https://www.youtube.com/shorts/_P63NrBN49U",
"duration_text": null,
"view_count": 64000000,
"published_time": "1 year ago",
"channel": { "id": null, "name": "", "url": null }
}
],
"next_cursor": "GhkKF...",
"has_more": true
},
"response_time": 2290,
"credits_used": 2,
"credits_remaining": 4809
}Every field is structured and typed -- no HTML parsing, no CSS selectors, no regex extraction. Your TypeScript code can access any field directly.
Step 4: Full Working Example
Here is a complete, runnable TypeScript script that searches YouTube Shorts and prints the results:
/**
* Search YouTube Shorts data with the Scavio API.
* POST /api/v1/youtube/shorts - rows come back under data.results, 2 credits per call.
* Run with: npx tsx youtube-shorts.ts
*/
// Shorts search takes search (not query). channel.name is often empty on Shorts rows -
// resolve it with POST /api/v1/youtube/video when you need it.
const API_URL = "https://api.scavio.dev/api/v1/youtube/shorts";
const API_KEY = process.env.SCAVIO_API_KEY as string;
interface YouTubeShortsResponse {
data: { results: Array<{ video_id: string; title: string; url: string; duration_text: string | null; view_count: number; published_time: string; channel: { id: string | null; name: string; url: string | null } }>; next_cursor: string | null; has_more: boolean };
response_time: number;
credits_used: number;
credits_remaining: number;
}
async function searchYouTubeShorts(search: string): Promise<YouTubeShortsResponse> {
const response = await fetch(API_URL, {
method: "POST",
headers: {
"Authorization": "Bearer " + API_KEY,
"Content-Type": "application/json",
},
body: JSON.stringify({ search }),
});
if (!response.ok) {
throw new Error("Scavio API error: " + response.status);
}
return (await response.json()) as YouTubeShortsResponse;
}
const data = await searchYouTubeShorts("ai coding agent shorts");
const rows = data?.data?.results ?? [];
for (const row of rows.slice(0, 5)) {
console.log(row?.title);
console.log(" ", row?.view_count, row?.url, row?.video_id);
}Why Use Scavio Instead of Scraping YouTube Shorts Directly?
- No proxy management. Direct scraping requires rotating proxies to avoid IP bans. Scavio handles all of this server-side.
- No CAPTCHA solving. YouTube Shorts aggressively blocks automated requests. Scavio returns clean data every time.
- Structured JSON output. No HTML parsing or CSS selector maintenance. Get typed, consistent data from every request.
- Multi-platform in one API. Search Google, Amazon, YouTube, and Walmart from the same API key with the same authentication pattern.
- Free tier included. 50 credits on signup with no credit card required. Each search costs 1 credit.