YouTube Video Scraper — No API Key, No Quota Limits

Search YouTube by keyword and get clean, de-duplicated video JSON — title, channel, views, duration, upload date. No API key, no quota. Free to start, MCP ready.

Runs with one click — no required fields. Type a query (or use the default), get clean, de-duplicated JSON. Built for both humans and AI agents (MCP-ready).

What it does

The official YouTube Data API v3 charges quota per call and caps you at 10,000 units/day — enough for a handful of searches before you're locked out. This actor reads YouTube's own public search results page (the same data your browser renders, no login) and returns every video as clean, normalized JSON: no API key to request, no quota to run out of, no Google Cloud project to set up. You give it a search query, it gives you the matching videos — de-duplicated and ready for your dashboard, spreadsheet, or AI agent.

Quick start (no code)

1. Click Start with the default query, or type your own in query.
2. Wait a few seconds — the actor searches and returns results as a dataset.
3. Export the dataset as JSON, CSV, or Excel directly from the Apify Console, or pull it via API.

Input

Field	Required	Description
query	no	What to search on YouTube (default: greenhouse jobs api)
maxResults	no	Max clean results (default 20, cap 500)
enrich	no	Deterministic enrichment per record (domain extraction, completeness score)
monitor	no	Compare with the previous run, flag NEW videos only — ideal for scheduled runs tracking a topic over time

Output (dataset, per record)

Real example — query "greenhouse jobs api" (captured live 2026-08-11, unedited):

{
  "id": "TvZvYx2fGHw",
  "title": "Building on Your ATS: Greenhouse API + Make Demo",
  "url": "https://www.youtube.com/watch?v=TvZvYx2fGHw",
  "channel": "PromptMates",
  "channel_url": "https://www.youtube.com/@PromptMates",
  "views_text": "710 views",
  "duration": "14:54",
  "published_text": "5 months ago",
  "description": "All our resources (community, live sessions, newsletter)...",
  "thumbnail": "https://i.ytimg.com/vi/TvZvYx2fGHw/hq720.jpg",
  "domain": "www.youtube.com",
  "completeness": 0.33
}

The final dataset row _summary carries run totals (total_clean, deduped, enriched). In monitor mode, a _changes row lists the video IDs that are new since the last run.

Use cases

• Content research — see what's already ranking for a topic before you script your own video.
• Competitor tracking — monitor a niche's search results weekly (monitor: true) to catch new uploads the moment they appear.
• Market/trend signals — video titles and upload cadence for a keyword are a cheap, fast proxy for what's heating up.
• Lead-gen adjacent — find channels actively publishing about a product/integration (e.g. companies with tutorial videos about a competitor's API).

Enrichment (optional, charged only when it produces something)

With enrich: true, each record gets a domain field (canonical host extracted from the video URL) and a completeness score (0–1). No LLM is used anywhere in this actor — enrichment is deterministic, so output is stable run to run and costs are fully predictable in advance.

Monitor mode — change alerts on a schedule

Set monitor: true and schedule the actor (daily/weekly via Apify's built-in Scheduler). Each run compares the current result set against the previous one and adds a _changes row listing only the video IDs that weren't there before — so you can wire an alert (email, Slack, Telegram) off new uploads for a topic without re-processing the whole result set yourself.

Use it from your code

from apify_client import ApifyClient

client = ApifyClient("<YOUR_API_TOKEN>")
run = client.actor("enezli/youtube-scraper").call(run_input={
    "query": "greenhouse jobs api",
    "maxResults": 50,
    "enrich": True,
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item)

curl "https://api.apify.com/v2/acts/enezli~youtube-scraper/run-sync-get-dataset-items?token=<YOUR_API_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{"query": "greenhouse jobs api", "maxResults": 20}'

Use it with AI agents (MCP)

This actor is listed under the MCP_SERVERS category and works out of the box as an Apify MCP tool — an agent can call it directly with a query and get structured JSON back, no scraping code for the agent to write or maintain.

Pricing — Pay-Per-Event, start is free

You only pay for what you get — no subscription, no charge for a run that finds nothing useful:

Event	When charged
actor-start	$0.0001 — symbolic, effectively free
result	$0.002 per clean video record returned
enrichment	$0.01 — only for a record that actually got enriched
change-alert	$0.005 — monitor mode only, per NEW video since the previous run

Volume discounts apply automatically via Apify account tiers (up to −44% on GOLD).

Is this legal?

This actor collects publicly available data only — the same search results any visitor sees in a browser, without logging in. It does not bypass authentication, does not access private videos, and does not store personal credentials. You are responsible for using the output in compliance with YouTube's Terms of Service and applicable law (e.g. GDPR) in your jurisdiction.

Support & feedback

Found a bug, or need a field this actor doesn't return yet (e.g. like/subscriber counts — see Limitations below)? Open an issue on the actor's Issues tab. Feature requests are welcome; frequently-requested fields get added to the standard output in a future version.

Changelog

• 0.1 (2026-08-11) — Initial release: keyword search, clean/de-duplicated JSON, optional deterministic enrichment, monitor mode.

Limitations (honest ones)

• No like or subscriber counts in this version. Those live on the individual video/channel page, not the search-results page this actor reads — adding them means an extra request per video, planned for a future version, not silently promised here.
• YouTube's search page markup can change without notice; if a run suddenly returns 0 results, please open an issue so the parser can be updated.
• This is search-result data, not the full YouTube Data API surface (no comments, no captions in this version).

FAQ

Does this need a YouTube API key? No — it reads the public search page directly, no Google Cloud project or API key required. Will I get rate-limited? The actor makes one request per run for up to ~20 results; for larger maxResults it's still a small, polite number of requests. No login means no per-account YouTube quota applies. Can I search by channel instead of keyword? Not in this version — this actor is keyword-search only. A dedicated channel-scraper variant is planned. Why is completeness sometimes low? The scoring field was designed for e-commerce/business data (price, location, rating); most of those don't apply to video content, so a "low" score here is expected and not a data-quality problem — check the YouTube-specific fields (channel, views_text, duration) instead.