Telegram Channel Scraper + Keyword Search
Pricing
from $1.50 / 1,000 posts
Telegram Channel Scraper + Keyword Search
Scrape posts from public Telegram channels without a login. Search a channel's whole history by keyword, filter by date range, and schedule runs that return only new posts.
Pricing
from $1.50 / 1,000 posts
Rating
0.0
(0)
Developer
Luka Puhmeister
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
15 hours ago
Last modified
Categories
Features
Share
Get posts from public Telegram channels as JSON, CSV or Excel. No login, no phone number, no Telegram account.
- Search by keyword inside a channel. Uses Telegram's own search across the channel's whole history, with no Google dependency and no extra search costs.
- Complete mode. Reads every post in a date range and matches keywords itself, for when Telegram's search stops early.
- Monitor mode with no bookkeeping. Schedule it and each run returns only new posts. The Actor remembers where it stopped, per channel and keyword set. You never pass a cursor.
- Both ends of a date range. Set a start, an end, or both. An old date window jumps straight there instead of reading years of posts.
What data does it return?
| Field | Example | Notes |
|---|---|---|
id | durov/551 | Stable, use it to de-duplicate |
channel | durov | The username Telegram serves the post under |
requestedChannel | durov | What you asked for. Differs when a channel has several usernames: breakingmash serves posts as mash |
messageId | 551 | |
url | https://t.me/durov/551 | |
date | 2026-10-10T14:22:43+00:00 | |
text | 💸 If your wallet... | Line breaks and emoji kept |
views, viewsRaw | 2040000, 2.04M | Telegram rounds big numbers |
author | Pavel Durov | The post signature, when the channel uses them |
isEdited | false | |
isForwarded, forwardedFrom | true, { "name": "Telegram Designers", "url": "https://t.me/designers/266", "author": null } | name is empty when the source is hidden |
replyTo | { "url": "...", "author": "...", "text": "..." } | |
mediaType | video | photo, video, album, round_video, voice, audio, document, sticker, poll, unsupported or empty for plain text |
photos | ["https://cdn4.telesco.pe/..."] | All photos of an album |
video, videos | { "url": "...", "thumbnail": "...", "duration": "1:15" } | video is the first one, videos lists all. url is empty when Telegram shows "Media is too big" |
document | { "title": "...", "extra": "..." } | Title only, Telegram does not expose the file |
voice | { "url": "...", "duration": "0:03" } | |
poll | { "question": "...", "type": "Anonymous Poll", "voters": 21400, "options": [{ "text": "...", "percent": 13 }] } | |
linkPreview | { "url": "...", "siteName": "...", "title": "...", "description": "...", "image": "..." } | |
reactions | [{ "emoji": "🤡", "emojiId": null, "count": 1420, "countRaw": "1.42K" }] | Custom emoji have only an emojiId, the page does not show a character for them |
reactionsTotal | 61260 | Sum of reactions, without paid stars |
paidStars | 5 | |
hashtags, mentions, links | ["AI"], ["durov"], ["https://..."] | |
matchedKeywords | ["wallet"] | Empty when you did not set keywords |
channelTitle, channelSubscribers | Pavel Durov, 10100000 | |
scrapedAt | 2026-11-04T12:00:00+00:00 |
Media URLs carry tokens and expire. Download the files you need soon after the run.
How much does it cost?
Pay per event, with no subscription and no platform usage on top:
| Event | Price |
|---|---|
| Run start | $0.005 |
| Channel read | $0.002 per channel |
| Post returned | $0.0015 per post ($1.50 per 1,000) |
So 1,000 posts from one channel cost $1.507. A channel with no public preview costs nothing and is never billed as a result.
Output example
One item from a real run, long texts shortened:
{"id": "durov/552","channel": "durov","messageId": 552,"url": "https://t.me/durov/552","date": "2026-10-11T14:52:40+00:00","text": "✨ Designers × AI\n\n🆕 The mockup era is over. Today’s AI tools give every designer a superpower: you can now craft...","views": 2670000,"viewsRaw": "2.67M","author": "Pavel Durov","isEdited": false,"isForwarded": true,"forwardedFrom": {"name": "Telegram Designers","url": "https://t.me/designers/266","author": null},"replyTo": null,"mediaType": "video","photos": [],"video": {"url": null,"thumbnail": "https://cdn4.telesco.pe/file/E22fRJ-zY1kHBKt8IFVngVgh694n_petLTWDG_O2BDO9ItJyagQZntBhYBMi2OtjeNjZiXY09nD_NYw6V4B4ubOg...","duration": "1:15"},"videos": [{"url": null,"thumbnail": "https://cdn4.telesco.pe/file/E22fRJ-zY1kHBKt8IFVngVgh694n_petLTWDG_O2BDO9ItJyagQZntBhYBMi2OtjeNjZiXY09nD_NYw6V4B4ubOg...","duration": "1:15"}],"document": null,"voice": null,"poll": null,"linkPreview": null,"reactions": [{"emoji": null,"emojiId": "5936157098181135162","count": 20400,"countRaw": "20.4K"},{"emoji": null,"emojiId": "5399847211989246390","count": 8680,"countRaw": "8.68K"},{"emoji": null,"emojiId": "5373223594484587136","count": 6880,"countRaw": "6.88K"}],"reactionsTotal": 35960,"paidStars": 10900,"hashtags": [],"mentions": ["designers"],"links": [],"matchedKeywords": [],"scrapedAt": "2026-11-09T22:30:16+00:00","channelTitle": "Pavel Durov","channelSubscribers": 10100000,"requestedChannel": "durov"}
How to use it
- Add Channels: a username, an @name or a t.me link.
- Optional: add Keywords, a date range, or turn on Only new posts.
- Set Max posts per channel so the run stops where you want it to.
- Run, then download from the Dataset tab or pull results through the API.
Input example
{"channels": ["durov", "https://t.me/telegram"],"keywords": ["wallet"],"searchMode": "telegram","dateFrom": "2025-01-01","dateTo": "2026-06-30","maxPostsPerChannel": 200,"onlyNew": false}
Telegram search or complete mode?
Telegram search (the default) is fast and reaches the whole history, because Telegram does the searching. It has two quirks. For very common words it can stop early: telegram in @durov returned 22 posts out of hundreds. It can also return a post where the word is not visible.
Complete mode reads every post and matches keywords at the start of a word, in the text and the link preview: wallet finds wallets, and art does not find start. It misses nothing but reads more pages, so use it with a date range. It stops after Max pages to read per channel (1,000 by default, about 20,000 posts) and notes that in the run's report.
Monitor mode
Turn on Only new posts and put the Actor on a schedule.
- The first run returns the latest posts and sets the starting point.
- Later runs return only what is new, oldest first.
- If there are more new posts than Max posts per channel, or your Max total charge runs out, the rest comes in the next run. Nothing is skipped.
- Each channel and keyword set has its own position, kept in a key-value store named
telegram-channel-scraper-statein your account. Delete a record there to start that channel over.
A run with nothing new still pays the run start and channel events.
Use it from code
from apify_client import ApifyClientclient = ApifyClient("YOUR_API_TOKEN")run = client.actor("PuhmeisterLuka/telegram-channel-scraper").call(run_input={"channels": ["durov"], "keywords": ["wallet"], "maxPostsPerChannel": 100})for item in client.dataset(run["defaultDatasetId"]).iterate_items():print(item["url"], item["text"])
Works the same from Node, Make, Zapier, n8n, or as an MCP tool for AI agents.
FAQ
Why is a channel missing from my results?
Some channels turn off the public web preview, and @bbcnews is one. Those cannot be read without an account, so the Actor skips them and does not bill them. The run's status message names them, and the CHANNELS record in the run's key-value store has a line per channel with its status, subscriber count, description and post count.
Why did I get fewer posts than Max posts per channel? The channel or the date range ran out, or your Max total charge was reached. You are only billed for posts delivered.
Why are some posts empty?
Telegram's web page shows "Please open Telegram to view this post" for some content. Those come back with mediaType: "unsupported", plus their date, views and reactions. Service messages such as "Channel created" are skipped.
Can it read private channels or groups? No. Public pages only, with no login.
Does it need a proxy? No. Leave it off unless you see blocks.
Is this legal? The Actor reads only pages anyone can open without an account. Check Telegram's terms and your local law for how you use the data, especially personal data under GDPR.
Something broke? Open an issue on the Issues tab. Fixes usually ship within a day.
More scrapers
Job boards in India, the Netherlands, Germany, Switzerland, Denmark, Poland, Singapore and the US, Remote OK, and any company's Ashby, Lever, Greenhouse, SmartRecruiters or ADP board: all Actors by this developer.