Telegram Keyword Search Scraper: Posts From Public Channels avatar

Telegram Keyword Search Scraper: Posts From Public Channels

Pricing

from $1.35 / 1,000 posts

Go to Apify Store
Telegram Keyword Search Scraper: Posts From Public Channels

Telegram Keyword Search Scraper: Posts From Public Channels

Search public Telegram channel posts by keyword: post text, channel, subscribers, date, views, reactions, forwarded from, links and post URL. Finds the channels for you or searches your list. Only posts holding your words are charged. No login, phone number or API key.

Pricing

from $1.35 / 1,000 posts

Rating

0.0

(0)

Developer

The Mine Works

The Mine Works

Maintained by Community

Actor stats

0

Bookmarked

1

Total users

0

Monthly active users

6 days ago

Last modified

Features

Limited permissions

Share

30 Telegram posts about bitcoin etf with views and links in 19 seconds

From The Mine Works, makers of Telegram Channel Scraper and Telegram Channel Finder, with over 290,000 runs across 310+ public actors.

Updated 4 Nov 2026: first release. Keyword search across public Telegram channels, your own channel list, date, views and subscriber filters, deep scan of older posts, and a monitor mode for schedules. Bookmark this Actor on its Store page so it is one click away next time.

Why choose this actor?

  • Every row is a post that holds your words. A post is read and checked before it is delivered, not just the latest post of a page that mentioned your topic. Telegram's own search is loose: in our test, "remote python jobs" brought back 78 posts and only 26 held all three words. Posts without your words are dropped and never charged.
  • The channels are found for you, and you can add your own. Give a keyword and the actor finds public channels that talk about it, on the web and from the channels that matching posts were forwarded from. List channels you already know and they are searched first.
  • A search that finds nothing costs nothing. No start fee and no per-keyword fee on an empty run. You pay per post, plus a small flat fee only on runs that delivered a post. No Telegram account, phone number, API key or cookies.

Run it on Apify

Part of The Mine Works Social media and video family: Threads Scraper, Reddit Scraper, Threads Search Scraper, Instagram Profile Scraper, Instagram Followers & Following, Reddit Search Scraper.

What this actor can and cannot see

Telegram has no public search across all channels. Its app can do more, but that needs an account, and this actor never logs in. So it works in two steps, and you should know the limit of each:

  1. It reads only channels it has found. Channels come from web search results, a public index of Telegram posts, and forwards. A channel none of these name is not read. In a spot check on two keywords, 6 of 20 active channels we knew carried the keyword were found (4 of 10 for "bitcoin etf", 2 of 10 for "openai"). Treat results as a strong sample of the conversation, not a census. If certain channels must be covered, list them in channels: a listed channel is always searched.
  2. Telegram answers about 22 newest matches per channel and keyword. To go further back, set deepScanPages, which reads the channel's history page by page and matches it here.

Private channels, private groups, direct messages and channels whose owner turned off the web preview cannot be read. About 1 in 4 names found on the web had no preview; they are skipped and never charged.

Try it in one minute

Paste this into the input (JSON view) and press Start. It returns 30 posts about "bitcoin etf". It took 19 seconds in our test run:

{
"keywords": ["bitcoin etf"],
"maxItems": 30
}

Put several keywords on separate lines to run several searches at once. A post that two of your keywords find is delivered and charged once.

Apify's free plan includes $5 of credit every month, which covers about 2,200 posts at this actor's Free tier price.

Copy to your AI assistant

themineworks/telegram-keyword-search-scraper on Apify. Searches public Telegram channel posts by keyword with no login and returns one flat row per post: text, channel, subscribers, date, views, reactions, forwarded from, links and post URL. Telegram has no public search across all channels, so it finds channels on the web, or uses a list you give, and searches inside each (about 22 newest matches per channel). Call ApifyClient("TOKEN").actor("themineworks/telegram-keyword-search-scraper").call(run_input={"keywords": ["bitcoin etf"], "maxItems": 100}), then client.dataset(run["defaultDatasetId"]).list_items().items. Rows with a _type field are summary, info or error rows. Optional: channels, discoverChannels, postedAfter ("7 days"), minViews, minSubscribers, deepScanPages, monitorMode.

Key features

  • Real matches, checked word by word. Each post, its link preview and its file name are checked for every word of your keyword. Word endings are allowed for, so "нефть" also finds "нефти". Each row says so in keyword_in_text. Turn requireAllWords off to take everything Telegram returned.
  • Search syntax. Plain words are all required, "spot etf" in quotes needs that exact phrase, #airdrop searches a hashtag and -scam drops posts holding that word. Any language works; our test keywords included Russian and Chinese.
  • Channel context on every row. Subscribers, title, bio and verified badge sit next to each post, so a mention can be weighed by reach.
  • Filters before billing. postedAfter, postedBefore, minViews and minSubscribers drop posts and channels before you pay for them.
  • Deep scan for older posts. deepScanPages walks a channel's history, 20 posts a page, and matches it here. Our test reached posts well past Telegram's newest 22 at the same price per post.
  • Monitor mode. On a schedule, monitorMode delivers and charges only posts your account has not received from this actor before.
  • Stops cleanly. Eight opened channels in a row with no match end a keyword, a page gets two tries at most, and a blocked run stops within four requests. The summary row says why a run ended.

How to use it

Basic: what is being said about a topic

{
"keywords": ["lab grown diamonds"],
"maxItems": 100
}

Sort by channel_subscribers or views to see which mentions reached people.

Brand monitoring on a schedule

{
"keywords": ["your brand", "\"your product name\""],
"postedAfter": "7 days",
"monitorMode": true,
"maxItems": 200
}

Save it as a task and schedule it daily. Only new posts arrive and only new posts are charged.

Search only the channels you follow

{
"keywords": ["airdrop", "listing"],
"channels": ["cointelegraph", "@WatcherGuru", "https://t.me/bitcoinnews"],
"discoverChannels": false
}

No web search is used. Each listed channel is asked for each keyword.

Reach further back in known channels

{
"keywords": ["telegram"],
"channels": ["durov", "telegram", "tginfo"],
"discoverChannels": false,
"deepScanPages": 10,
"maxPostsPerChannel": 100,
"postedAfter": "3 years"
}

Deep scan reads up to 10 extra history pages per channel. In a shorter test with 3 pages, 47 of 100 posts came from the history scan.

Larger channels only

{
"keywords": ["remote python jobs"],
"minSubscribers": 1000,
"minViews": 100
}

A simple way to leave out small and spam channels. Some search results are low quality channels, so this filter is worth using for commercial keywords.

Input parameters

ParameterTypeDefaultWhat it does
keywordsarray of stringsrequiredWhat to search for. Up to 100 keywords per run.
channelsarray of stringsemptyPublic channels to search first: durov, @durov or https://t.me/durov.
discoverChannelsbooleantrueFind channels on the web. Off: only your list is searched.
maxItemsinteger100Most posts per keyword, 1 to 10,000.
maxChannelsPerKeywordinteger50Most channels opened per keyword, your list included.
maxPostsPerChannelinteger20Most posts from one channel per keyword. Above 22 needs deep scan.
deepScanPagesinteger0History pages per channel to read and match, 20 posts a page.
postedAfterstringempty2026-10-01, or 7 days, 2 weeks, 1 month back from today.
postedBeforestringemptySame forms; keeps older posts.
minViewsinteger0Drop posts with fewer views.
minSubscribersinteger0Skip channels with fewer subscribers.
requireAllWordsbooleantrueOnly posts holding every word. Off: all Telegram returned, flagged in keyword_in_text.
monitorModebooleanfalseDeliver and charge only posts not delivered to your account before.

What data do you get?

One row per post. Post rows have no _type field; the summary, info and error rows do, so a filter on _type separates them.

FieldWhat it is
post_url, post_idDirect link to the post, and its number in the channel
keyword, keyword_in_textThe keyword that found it, and whether every word is in the post
textFull post text
posted_at, editedWhen it was posted (UTC), and whether it was edited
views, views_labelViews as a number, and as Telegram shows them (6.79K). Telegram rounds large counts
reactions, reactions_totalReactions by emoji, and their sum
forwarded_from, forwarded_from_urlWhere a forwarded post came from
reply_to_url, authorThe post it replies to, and the author signature when the channel signs posts
linksOutbound links in the text and link preview; Telegram's own links are left out
link_preview_url, link_preview_title, link_preview_description, link_preview_siteThe preview card
hashtags, mentions, emailsHashtags, usernames and email addresses in the post
has_media, media_types, photo_urls, video_urls, document_titleMedia in the post. Media links expire after a while
channel, channel_url, channel_titleThe channel
channel_subscribers, channel_verified, channel_descriptionIts reach, badge and bio
discovered_viaHow the channel was reached: your_list, duckduckgo, lyzem, forwarded_from or google
match_sourcetelegram_search or history_scan
scraped_atWhen the row was written

Forward counts are not included. Telegram's public web preview does not show how many times a post was forwarded, only where a forwarded post came from, which is in forwarded_from.

Summary row (_type: "summary", never charged): keywords_requested, keywords_read, channels_opened, channels_with_matches, channels_no_preview, channels_blocked, posts_seen, posts_without_all_words, posts_filtered_out, posts_already_delivered, duplicates_skipped, posts_delivered, charged_for, http_requests, stopped_by_max_cost, stopped_by_block_guard, ended_on_deadline, shortfall_reason when fewer posts came back than asked, and per_keyword, one entry per keyword with channels found and opened, posts delivered, what each source gave and why it stopped. The same summary is saved as the OUTPUT record of the run's key-value store.

Error rows (_type: "error", never charged): one per input that could not be read, with error_type (invalid_channel, invalid_date), identifier and reason.

Stable fields for automations

These were present in every post row of our test runs, and their names will not change:

FieldWhat it is
post_urlThe key to store and join on
post_idPost number in the channel
keywordSearch that found it
keyword_in_textEvery word present
textPost text
posted_atDate and time, UTC
viewsView count
channelChannel username
channel_urlChannel link
channel_titleChannel name
channel_subscribersSubscribers
channel_verifiedVerified badge
has_mediaHas media
discovered_viaHow the channel was reached
match_sourceHow the post was matched
scraped_atWhen the row was written

Optional, present only when the post has a value. In a test of 1,640 posts: links 952, reactions 844, media_types 801, photo_urls 707, hashtags 451, link_preview_url 336, mentions 186, author 84, forwarded_from 75, video_urls 52, emails 35, reply_to_url 26, document_title 21.

Output examples

One post from a "bitcoin etf" search (test run, 4 Nov 2026):

{
"post_url": "https://t.me/WatcherGuru/13287",
"post_id": 13287,
"keyword": "bitcoin etf",
"keyword_in_text": true,
"text": "JUST IN: $3.5 trillion Goldman Sachs files for Bitcoin Premium Income ETF.\n\n📰 Full Story\n\n@WatcherGuru",
"posted_at": "2026-04-14T15:11:36+00:00",
"edited": false,
"views": 123000,
"views_label": "123K",
"reactions": [
{ "emoji": "🔥", "count": 664 },
{ "emoji": "❤", "count": 215 },
{ "emoji": "💩", "count": 102 },
{ "emoji": "🤡", "count": 63 },
{ "emoji": "🤯", "count": 19 }
],
"reactions_total": 1063,
"links": ["https://watcher.guru/news/goldman-sachs-files-for-bitcoin-premium-income-etf"],
"mentions": ["WatcherGuru"],
"has_media": true,
"media_types": ["photo"],
"channel": "WatcherGuru",
"channel_url": "https://t.me/WatcherGuru",
"channel_title": "Watcher Guru",
"channel_subscribers": 631000,
"channel_verified": true,
"discovered_via": "duckduckgo",
"match_source": "telegram_search",
"scraped_at": "2026-11-04T14:06:56.094Z"
}

The summary row of the "Try it" run, trimmed:

{
"_type": "summary",
"keywords_requested": 1,
"channels_opened": 5,
"channels_with_matches": 5,
"posts_seen": 101,
"posts_without_all_words": 1,
"posts_delivered": 30,
"http_requests": 6,
"http_blocked": 0,
"per_keyword": [
{ "keyword": "bitcoin etf", "channels_found": 7, "channels_opened": 5, "posts_delivered": 30, "stopped": "max_items_reached" }
],
"run_seconds": 19
}

Pricing

Pay per event: you pay for the posts you receive, plus a flat $0.003 on runs that delivered a post, whatever memory you choose. Finding channels costs nothing extra, and neither does deep scan.

EventWhen it is chargedFree planBronze (Starter)Silver (Scale)Gold (Business)
post-foundOnce per post row stored in your dataset$2.25 per 1,000$1.90 per 1,000$1.60 per 1,000$1.35 per 1,000
run-startOnce per run, after the first post was delivered$0.003$0.003$0.003$0.003

The price of one post is the price per 1,000 divided by 1,000.

Never charged:

  • posts that do not hold your words (with requireAllWords on, the default);
  • posts and channels your filters leave out;
  • a post that a second keyword of the same run finds again;
  • in monitor mode, posts your account already received;
  • channels with no preview, private chats and invite links;
  • the summary, info and error rows;
  • a run that delivered nothing: no start fee, no per-keyword fee;
  • anything past your "Max cost per run". The actor checks the budget before each batch of rows and stops while the total is still below your limit, so the run ends SUCCEEDED with a partial result and a status message saying so.

Run it on a schedule

  1. Open the actor in Apify Console, set your input and click Save as a new task.
  2. Add "monitorMode": true so each run delivers and charges only new posts.
  3. In the task, open Schedules, click Add schedule and pick how often (every hour, every Monday at 08:00 and so on).
  4. Optional: add an integration (Google Sheets, Slack, a webhook) so each run lands where your team works.

FAQ

Does this search all of Telegram?

No. Telegram offers no public search across channels without an account. This actor finds channels first, then searches inside each, so a channel it did not find is not covered. See "What this actor can and cannot see" above. To be sure of a channel, list it in channels.

How many posts can I get?

Up to 10,000 per keyword and 100 keywords per run, as far as the channels found allow. Telegram answers about 22 newest matches per channel and keyword. In our test of 10 keywords with a limit of 200 each, 7 reached the 200 and the narrow ones returned 54 to 95. More channels in channels, deepScanPages or a broader keyword bring more.

Why are some posts old?

Telegram returns each channel's newest matches, and in a quiet channel those can be years old. Of 1,640 test posts, 478 were from the last 30 days and 1,105 from the last year. Use postedAfter to keep only recent ones; older posts are then not charged.

Why did a channel I expected not appear?

It may not be named by any source, it may have no web preview (its owner can turn this off), or its posts may not hold every word. The summary's per_keyword shows what was found and opened. List the channel in channels to search it directly.

Are forward counts available?

No. Telegram's web preview does not publish them. forwarded_from shows where a forwarded post came from. Views and reactions are the reach numbers available.

Are the results clean of spam?

Not fully. Public channels include promotions and spam, and they can hold your words like anything else. minSubscribers and minViews remove most of it, and discovered_via shows where a channel came from.

What is the difference from Telegram Channel Scraper and Telegram Channel Finder?

Telegram Channel Finder returns channels for a keyword, not posts. Telegram Channel Scraper returns everything a known channel posted. This actor returns the posts that hold your keyword, across channels.

How long does a run take?

In our tests: 19 seconds for the "Try it" input, 24 seconds for 100 posts from 3 listed channels with deep scan, about 80 seconds per keyword for 200 posts when the web search has to dig deep. An empty keyword ends in about 5 seconds. The default time limit is 3,600 seconds; the actor stops before it and keeps every row.

Do I need an account, phone number or proxy?

No. The actor reads public web pages over Apify Proxy, which is included. Nothing to set up.

Which formats can I download?

JSON, CSV, Excel, XML, RSS and HTML from the run's Storage tab or through the API. The dataset has three ready views: Posts, Reach and engagement, and Links and contacts.

Can I use it from Claude, ChatGPT or Cursor (MCP)?

Yes, through Apify's MCP server. Connector URL: https://mcp.apify.com/?tools=themineworks/telegram-keyword-search-scraper

  • Claude: Settings > Connectors > Add custom connector, paste the URL, sign in with Apify.
  • ChatGPT: turn on developer mode, add an MCP connector with the URL, sign in with Apify.
  • Cursor or VS Code: add it as an HTTP MCP server with that URL.
  • Claude Code: claude mcp add -t http telegram-keyword-search-scraper "https://mcp.apify.com/?tools=themineworks/telegram-keyword-search-scraper"

The actor reads only public channel previews that anyone can open in a browser without an account. It does not read private chats or log in. Posts can contain personal data, so you are responsible for how you use the data, including Telegram's terms and data protection law such as GDPR and CCPA. This is general information, not legal advice.

Integrations

  • Google Sheets: send each run's posts to a sheet with Apify's Google Sheets integration; one row per post fits a sheet as is.
  • Make, Zapier and n8n: start a run and read its dataset with the Apify apps and nodes, for example "new posts naming our brand, every morning, to Slack".
  • Webhooks: have Apify call your URL when a run succeeds, then fetch the dataset.
  • API and SDKs: run it from Python or JavaScript with the Apify client, or with a plain HTTP call (see "Copy to your AI assistant" above).
  • MCP clients: Claude, ChatGPT, Cursor and VS Code can run it as a tool through the connector URL in the FAQ.

More from The Mine Works

Social media and video

Leads and business directories

Marketing, SEO and reviews

LinkedIn

Real estate

Science, health and government data

Jobs and hiring

Company and business data

E-commerce and marketplaces

Food and local services

Developer and AI tools

More tools

Support

Found a keyword that returns fewer posts than you expected, or a field that comes back empty? Open an issue in the Issues tab with the input and the run ID, and we will look at it. For a field you need or a new source, email dmineworks@gmail.com.

Search public Telegram channels by keyword and get one row per post that holds your words, with text, channel, subscribers, date, views, reactions, forwarded from and links, no login.