Social Media · Web scrapers
YouTube Scraper API
- 13 endpoints
- $1.49 per 1,000 calls
- Structured JSON every response
- Errors and timeouts are not charged
For developers
One GET, and YouTube comes back as JSON.
This is the YouTube channel endpoint, $1.49 per 1,000 calls. Your key goes in one header; the answer is the object below it.
- Proxies, sessions and captchas handled on our side
- Failed calls are retried and not charged
- The same JSON shape on every call
curl "https://hproxy.com/api/v1/scrape/youtube/channel-lookup?handle=MrBeast" \
-H "X-API-Key: hpx_your_key_here"{
"id": "UC...",
"title": "Channel name",
"about": "What the channel is about.",
"subscriberCount": 412000,
"videoCount": 318,
"links": ["https://example.com"],
"banner": "https://.../banner.jpg"
}Endpoint catalog
13 YouTube scrapers, ready to call.
One REST call each, billed per call from 0.15¢, with structured JSON back every time.
What you scrape
Pull a YouTube channel in full: about text, country, subscriber, view and video counts, every external link and its artwork.
Pull every long-form video on a YouTube channel, page by page, with title, views, likes, duration and publish date.
Pull every Short on a YouTube channel, page by page, with title, view and like counts, duration and publish date.
Pull a single YouTube video from its URL: title, description, views, likes, comment count, channel and publish date.
Pull the full timestamped transcript of any YouTube video, every caption line with its start and end time, in any language.
Run any YouTube search and get the ranked results back as JSON: videos, channels and playlists in the order YouTube returns them.
Pull every comment on a YouTube video, page by page, with author, text, like count, reply count and timestamp.
Pull the Shorts trending on YouTube right now, recomputed live on every call, with view counts and the channel behind each.
Pull the videos published under any YouTube hashtag, page by page, with engagement stats and channel metadata.
Pull the full reply thread under any YouTube comment, with authors, like counts and timestamps at every level.
Pull a YouTube playlist in full: every video in playlist order with title, duration and channel, plus the playlist owner.
Pull a YouTube community post: text, attached image or poll, poll results where public, and reaction and comment counts.
The same data on every other platform: Profile Scraper API, Post Scraper API, Transcript API, Comment Scraper API, Search API.
How it works
Scraping YouTube, in three steps.
You write one line. Proxies, anti-bot walls, rendering and parsing happen on our side.
- 1
Create a free account
Sign up and top up any amount. The balance pays for every HProxy product. - 2
Grab your API key
One key from the dashboard authenticates every scraper. - 3
Call the endpoint
Send the GET you need and read the JSON. That is the entire integration.
Why it holds up
What you do not have to build.
No captchas, ever
Anti-bot walls, JS rendering and rate limits are handled server-side. You call an API, that is the whole job.
Residential-backed
Every request exits through HProxy's own residential network, so it looks like a real user on a real device.
Pay per call
No subscription, no seat licenses. Calls are billed from your balance.
Fresh on every call
Responses are scraped live at request time, no stale cache dressed up as data.
Errors are free
Blocked, errored or timed-out calls are not charged. A target that does not exist counts as a call, because the platform still had to be asked.
Scales with you
From one call to millions: same endpoint, same price curve, volume discounts as you grow.
- Pay per call from one balance, no subscription and no monthly minimum.
- Blocked, errored and timed-out calls are not charged.
- Proxies, retries and parsing are included in the price.
- The same endpoint from one call to millions.
Docs and AI
Connect your scraper to an AI, any time.
The whole Scraper API is published as markdown, so an assistant reads every endpoint, every parameter and a real response without a key and writes your integration. And because a scrape is one GET that answers in JSON, an agent can run the scraper itself, on its own schedule, with your key.
Claude, ChatGPT, Cursor, Gemini, Copilot, Windsurf, Perplexity, Ollama and any assistant that speaks MCP.
The API as markdown
Paste /llms/scraper-api.txt into Claude, ChatGPT, Cursor or Copilot: every endpoint and the whole catalog, in one file. No key and no account to read it.
An agent can call it
One GET with your key returns structured JSON, so an agent scrapes on its own schedule. No browser to drive, no parser to keep alive between runs.
MCP server and CLI
Built by us for the free tools and hosted at mcp.hproxy.com: the proxy list, the checker and the IP lookup as tools in Claude, ChatGPT, Cursor or any assistant that speaks MCP. Nothing to install, no key.
API docs
Every scraper endpoint documented end to end, with the parameters it takes and the response it really returns.
YouTube questions
YouTube scraping, answered.
You send one HTTPS request to the endpoint you need with your API key, and HProxy handles the rest server-side: proxies, sessions, retries and parsing. The response is clean, structured JSON. No headless browser, no captcha solving, no parser maintenance on your side.
Pricing starts at $1.49 per 1,000 calls (0.15¢ per call), billed from your HProxy balance. There is no subscription and no monthly minimum. You top up once.
The API only accesses publicly available data, the same pages anyone can open in a browser. You are responsible for using the data in line with applicable laws and YouTube's terms for your use case.
13 ready-made endpoints cover the public YouTube surface: profiles, content, engagement, search and more. Every endpoint returns the full structured object, not a stripped-down summary.
No. The scraper runs on HProxy's own residential network behind the scenes, so requests look like real users. You only make one API call, and the proxy layer is included in the per-call price.
Calls typically return in a few seconds. Errors are retried automatically, and unrecoverable errors and timeouts are not charged. A lookup whose target does not exist counts as a call, because the platform still had to be asked.
Every platform
116 scrapers, 23 platforms, one key.
The same REST call, the same JSON shape and the same balance behind all of them. Pick where you need data from.
Social Media
10 platforms, 86 endpoints
Streaming
4 platforms, 9 endpoints
Marketplace
2 platforms, 6 endpoints
Business / B2B
1 platform, 6 endpoints
Link in Bio
5 platforms, 5 endpoints
Ad Libraries
1 platform, 4 endpoints
More on the YouTube Scraper API
What the YouTube Scraper API is
The YouTube Scraper API is 13 ready-made endpoints behind one REST base URL: Channel, Channel Videos, Channel Shorts, Video, Video Transcript, Search, Video Comments, Trending Shorts, Hashtag Search, Comment Replies, Playlist and Community Post. Each one is a single GET with your API key, and each one answers with a structured JSON object rather than a page of HTML, so nothing on your side has to parse markup or keep a selector alive. The proxies, sessions, retries, rendering and anti-bot handling all run on our side, which is the part of a YouTube scraper that normally breaks.
What you can scrape from YouTube
The catalog covers 5 kinds of YouTube data: profiles (1 endpoint: Channel), posts and media (5 endpoints: Channel Videos, Channel Shorts, Video, Playlist and Community Post), transcripts (1 endpoint: Video Transcript), comments (2 endpoints: Video Comments and Comment Replies) and search and trends (3 endpoints: Search, Trending Shorts and Hashtag Search). Every endpoint returns the full object it describes, not a trimmed summary, and the JSON keeps the same shape from the first call to the millionth, so a script written today still parses next month.
What scraping YouTube costs
Calls are billed from your HProxy balance at 0.15¢ each, $1.49 per 1,000, with no subscription and no monthly minimum. Blocked, errored and timed-out calls are not charged. A lookup whose target does not exist does count as a call, because YouTube still had to be asked before we could tell you.
Scraping YouTube yourself
Writing the request is the easy part. What costs the time is everything around it: residential addresses that do not get blocked, sessions that survive a login wall, a captcha path, a headless browser for the parts that render in JavaScript, and a parser rewritten every time YouTube moves its markup. This API is that work, kept running by us, priced per call, and it answers in the same JSON whether you send one request a day or a hundred a second.
Ready when you are.
Your dashboard is ten seconds away. No sales call, no subscription, no minimum.