YouTube Community Posts Scraper With Keyword & Date Monitoring

YouTube Community Posts Scraper With Keyword & Date Monitoring

Monitor the Community tab of any YouTube channel and get back only the posts that matter. This scraper turns a one-off channel scrape into a scheduled monitor: run it on a timer with delta mode enabled, and each run returns only the new community posts published since the last run. Narrow further with a keyword filter, an approximate date range, and a country (geo) context.

Most community-post scrapers dump the whole feed every run, leaving you to diff it yourself. This one remembers what it already emitted per channel, persisted across scheduled runs, so a daily monitor genuinely returns a daily delta instead of the same 50 posts over and over.

What is YouTube Community Posts Scraper With Keyword & Date Monitoring?

It's an Apify actor that reads a YouTube channel's Community tab through a hybrid HTML + InnerTube API approach (Chrome TLS impersonation to avoid soft-blocks), and adds monitoring-specific features on top: deltaMode tracks every post ID it has already emitted per channel in a persistent Apify key-value store, so a scheduled run only pushes posts published since the last run — the very next identical run returns 0 new rows if nothing changed. A keywordFilter keeps only posts whose text contains a phrase, and dateFrom/dateTo bound posts by an approximate date derived from YouTube's own relative "2 weeks ago" style timestamps (YouTube exposes no exact date on Community posts, so this window is honestly labeled approximate).

What data can I extract with YouTube Community Posts Scraper With Keyword & Date Monitoring?

Field	Description
postId / channelId	Post's unique ID (used for delta-mode dedup) and the channel's ID.
author / authorUrl	Channel display name and canonical channel path.
text	Full post text (also the poll question for poll posts).
publishedTime	Relative published time exactly as YouTube shows it (e.g. "2 weeks ago").
likeCount / commentCount	Abbreviated count strings (e.g. "1.2K"), or null when not shown.
attachmentType	"video", "image", "poll", or null for text-only posts.
pollChoices	Poll answer options; empty array for non-poll posts.
images	Attached image URLs; empty array when none.
videoThumbnail	Thumbnail URL when the post attaches a video; null otherwise.
sourceUrl	The input channel URL this row came from.

Why not build this yourself?

YouTube's Community tab has no public API — this actor reads the same reverse-engineered ytInitialData + InnerTube browse continuation mechanism any scraper must use, but routes every request through Chrome TLS/JA3 impersonation and a soft-block detector (checking for a data-less shell, a consent interstitial, or "unusual traffic" wording) rather than trusting a 200 status code alone. The harder part to build yourself is the delta-tracking: persisting a per-channel seen-post-ID set across scheduled runs, correctly sized so the "recent window" scanned each run is neither too small (misses a post if the channel posted a lot since last run) nor unboundedly large (rescans the whole feed every time). This actor handles that state management for you.

One honest limitation: YouTube's Community posts show only a relative published time ("3 days ago"), never an exact timestamp — so dateFrom/dateTo filtering is necessarily approximate, off by up to a few days depending on when YouTube's own relative-time bucket rounds. This is disclosed in the input description rather than presented as exact filtering.

How to use data extracted from YouTube Community Posts Scraper With Keyword & Date Monitoring?

• Community managers — run a scheduled delta-mode monitor on your own channel (or a competitor's) to get alerted only on genuinely new posts, not a repeated full dump.
• Fan/news aggregation sites — pull new Community posts from a set of creators on a schedule and republish or summarize only what's new.
• Brand monitoring teams — use keywordFilter to watch for a product or campaign name mentioned in a creator's Community posts.
• Researchers — collect a dated archive of a channel's Community activity using dateFrom/dateTo for a specific historical window.

🔼 Input sample

Parameter	Required	Type	Description	Example
channelUrls	Yes	array	Channel URLs to monitor (@handle or /channel/UC...). One per line or a comma-separated string.	["https://www.youtube.com/@MrBeast"]
maxPosts	No	integer	Posts to return per channel (1-10000). In delta mode, this is the size of the recent window scanned each run — only unseen posts inside it are emitted. Default 10.	10
deltaMode	No	boolean	Track and skip previously-emitted post IDs (persisted per channel). First run returns all matches; the next identical run returns 0. Default false.	true
keywordFilter	No	string	Case-insensitive substring match against post text. Empty = keep every post.	"giveaway"
dateFrom / dateTo	No	string	ISO date bounds (YYYY-MM-DD). Common pitfall: YouTube only exposes relative published times, so these bounds are approximate — treat the boundary as fuzzy by a few days.	"2026-07-01"
countryCode	No	string	Two-letter code sent as gl in the InnerTube client context, fetching the feed as if from that country. Default "US".	"GB"
proxy	No	object	Optional. Falls back to Apify RESIDENTIAL automatically on a block and retries on a fresh IP. Default direct.	{"useApifyProxy": false}
requestTimeout	No	integer	Per-request timeout in seconds (1-300). Default 15.	15
maxRetries	No	integer	Retry attempts before switching proxy (1-10). Default 3.	3
retryDelay	No	integer	Initial backoff between retries in seconds, doubling each attempt (0-60). Default 2.	2

{
  "channelUrls": ["https://www.youtube.com/@MrBeast"],
  "deltaMode": true,
  "keywordFilter": "giveaway",
  "maxPosts": 20
}

🔽 Output sample

{
  "channelId": "UCX6OQ3DkcsbYNE6H8uQQuVA",
  "postId": "Ugkx1234567890abcdefg",
  "author": "MrBeast",
  "text": "New video dropping this Friday! Who's excited?",
  "publishedTime": "3 days ago",
  "likeCount": "142K",
  "commentCount": "8.1K",
  "attachmentType": "image",
  "images": ["https://yt3.ggpht.com/example.jpg"],
  "pollChoices": [],
  "sourceUrl": "https://www.youtube.com/@MrBeast"
}

▶️ Want to try other scrapers?

Actor	What it does
youtube-channel-keyword-research-scraper	Research channels by keyword and SEO signal.
youtube-video-details-comments-scraper	Get video details and comments.
youtube-search-scraper-channel-contact-links	Search YouTube and pull channel contact links.

How to extract YouTube Community posts programmatically

from apify_client import ApifyClient

client = ApifyClient("<YOUR_API_TOKEN>")
run = client.actor("SimpleAPI/youtube-community-posts-scraper-keyword-date-monitoring").call(run_input={
    "channelUrls": ["https://www.youtube.com/@MrBeast"],
    "deltaMode": True,
    "maxPosts": 20,
})

for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["author"], item["publishedTime"], item["text"][:60])

Every run's dataset can also be exported straight to JSON, CSV, Excel or an API endpoint from the Apify Console — no code needed for a one-off pull.

Is it legal to scrape YouTube Community posts?

Scraping publicly accessible data — including public Community tab posts — is generally lawful in the US under the precedent set in hiQ Labs v. LinkedIn, which held that scraping data not gated behind a login does not violate the Computer Fraud and Abuse Act. This actor reads only Community posts that are publicly visible with no login required. If you extract data tied to identifiable individuals, GDPR (EU) and CCPA (California) still govern what you do with it, so review your own use case.

❓ FAQ

How does deltaMode actually avoid duplicate posts? It stores every emitted post's ID per channel in a persistent Apify key-value store (community-seen-postids). Each run loads that set, skips any post ID already in it, and saves the updated set at the end — so a scheduled run only pushes posts that weren't there last time.

Why does the first delta-mode run return a lot of posts, and the next return zero? The first run has no seen-post history, so everything in the scanned window counts as new. If the channel hasn't posted anything since, the next run's scanned posts are all already in the seen set — 0 new rows is the correct, expected result, not a failure.

How accurate is dateFrom/dateTo filtering? Approximate. YouTube's Community tab only ever shows a relative published time like "2 weeks ago," never an exact timestamp — the actor converts that into an approximate calendar date, which can be off by a few days depending on where YouTube's own rounding falls.

What happens if maxPosts is set very high with deltaMode on? It sets how deep into the recent-posts window the actor scans each run — a very high value with deltaMode on means scanning further back each run, which is unnecessary once the channel's normal posting cadence is established; a value that safely exceeds posts-since-last-run is usually enough.

Does keywordFilter search the whole post or just a snippet? It matches against the post's full text (case-insensitive substring), including the question text for poll posts.

What if a channel's Community tab is empty or disabled? The actor logs that no posts were matched for that channel and moves on to the next one in channelUrls — it doesn't fail the whole run.

How is this different from other YouTube Community post scrapers on Apify? As observed on the Apify Store on 2026-07-26, the Community post scrapers we reviewed return a full snapshot every run with no built-in deduplication across scheduled runs, and none offer approximate date filtering derived from YouTube's own relative timestamps.

Conclusion

YouTube Community Posts Scraper With Keyword & Date Monitoring turns a full-feed scrape into a genuine scheduled monitor — new posts only, filtered by keyword and approximate date, with the seen-post state handled for you across runs.