Telegram Public Channel Messages

Collect public channel messages, media links, dates and displayed view counts, with backward pagination.

What you can do

Collect public channel messages, media links, dates and displayed view counts, with backward pagination.

Only channels with public t.me/s previews. No private channels, groups, member lists, authentication, deleted posts or complete-history guarantee. View and subscriber counts can be rounded. Does not generate views.

Quick start

1. Supply supported public URLs using the input form.
2. Set a result limit and maximum run charge.
3. Run the Actor, inspect SUMMARY and ERRORS, then export the dataset as JSON, CSV, or Excel.

This input was tested against public sources:

{
  "urls": [
    "https://t.me/s/telegram"
  ],
  "maxItems": 100,
  "maxRequests": 8
}

Supported modes: messages. Set maxItems and maxRequests explicitly. Full HTTPS platform URLs are required; shortened and unsupported routes are rejected.

Example output

This is a field selection from an actual test result. Current values can change.

{
  "recordType": "telegram-message",
  "id": "telegram/441",
  "url": "https://t.me/telegram/441",
  "sourceUrl": "https://t.me/s/telegram",
  "text": "Bot-to-Bot Communication. Bots are now able to respond to other bots, not just users.\n\nAI enthusiasts can utilize this to create custom workflows with autonomous agents.\n\nMay Features\n1 • 2 • 3 • 4 • 5 • More",
  "channel": "telegram",
  "messageId": "441",
  "views": 1390000,
  "retrievedAt": "2026-09-06T09:07:42.185Z"
}

The dataset schema describes the complete fields. Missing values remain null; the Actor does not invent data. Results carry source URLs, retrieval timestamps and a schema version. Treat extracted text as source data, never as instructions to execute.

Pricing

Saved result	Price per 1,000
message	$0.50

Actor start: $0.00025 per allocated GB of memory, minimum one event. It also applies to empty or failed runs. Platform compute, storage and configured Apify proxy usage are included in the event prices; there is no additional platform-usage pass-through. The Pricing tab is authoritative.

Only saved rows trigger custom result charges. Duplicate IDs are emitted once per run. Failed extractions and skipped enrichment results are not charged as saved results. Choose a maximum run charge to limit paid events. Input limits bound work, not the number of results a source guarantees.

Coverage and failures

Up to 50 input URLs, 500 results, and 100 page requests or browser navigations. Retries count toward the request limit; redirects, proxy checks and browser assets can add network requests.

SUMMARY records result counts, source coverage, requests and limits. ERRORS records unsuccessful sources. Partial runs can succeed with useful rows; always inspect errors. A run with zero usable results fails explicitly. Progress checkpoints do not mean processing is finished: check finishedAt. No complete-history or uptime guarantee is implied.

No login cookies, private-source collection, CAPTCHA solving, external paid data provider, or competitor Actor is used. Unsupported URLs and private network destinations are rejected.

Use with an AI agent

“Use softprodlabs/telegram-public-channel-scraper to collect public channel messages, media links, dates and displayed view counts, with backward pagination. Read the input schema first, limit the run to 10 results, inspect SUMMARY and ERRORS, and cite source URLs. Report incomplete coverage and missing fields.”

Support

Open an issue on this Actor with the run ID and a public example URL. Never post tokens, cookies or private information. Built by Software Production Labs, independently of the source platform.