Telegram Channel Scraper — Messages, Views & Media

Scrape messages from any public Telegram channel — no API key, no Telegram account, no proxy required. This Telegram channel scraper extracts message text, view counts, timestamps, media type, links, forward source, and reply threading from public channels accessible via t.me.

Use it for brand monitoring, crypto/news signal aggregation, competitive intelligence, or feeding Telegram content into LLM pipelines.

How the Telegram scraper works

Uses the public Telegram web preview at https://t.me/s/{channel} — the same URL anyone can open in a browser. Paginates backwards via ?before={message_id} to collect historical messages going as far back as the channel's public web history allows.

No unofficial MTProto API, no bot token, no phone number verification. Just HTTP + structural HTML parsing.

Why this Telegram scraper is more reliable than alternatives

Most Telegram scrapers use the MTProto API (requires a phone number and Telegram account) or depend on CSS class names that break on every Telegram web redesign. This actor uses structural XPath parsing on the semantic HTML structure — data-* attributes and element structure, not fragile class names. When Telegram updates its web UI, parse_confidence drops below 0.8 and warnings flag the affected fields — you get a drift signal, not silent wrong data.

Input

{
  "channels": ["telegram", "@durov", "https://t.me/cointelegraph"],
  "maxMessages": 200
}

Field	Type	Description
channels	string[]	Public channel usernames, @handles, or t.me URLs
maxMessages	integer	Max messages per channel (default: 100)
beforeId	integer	Optional cursor — only collect messages before this ID

Output

One row per message:

Field	Type	Description
channel	string	Channel username
message_id	integer	Telegram message ID
text	string|null	Message text content
views	string|null	View count (e.g. "1.5M", "42K")
datetime	string	ISO-8601 UTC timestamp
has_media	boolean	Whether the message contains media
media_type	string|null	"photo", "video", "document", "sticker", "audio", "poll"
links	string[]	HTTP/HTTPS links in message text
is_forwarded	boolean	Whether forwarded from another channel
forward_from	string|null	Source channel/user name if forwarded
reply_to	integer|null	Message ID being replied to
message_url	string	Full https://t.me/{channel}/{id} URL
parse_confidence	float	0.0–1.0 quality signal; below 0.8 check warnings
warnings	string[]	Parse quality warnings (e.g. "missing_datetime")
scraped_at	string	ISO-8601 UTC scrape timestamp

What it does NOT support

• Private channels — only public channels accessible via t.me/s/{channel}
• Comments/replies within posts — web preview shows channel messages only
• Media downloads — returns media type detection only, not URLs to media files
• Real-time streaming — pulls historical messages, not a live feed

Competitive edge

Feature	This actor	@maximedupre/telegram-channel-messages-scraper
parse_confidence field	✓	✗
Pagination cursor input (beforeId)	✓	✗
Forward source name	✓	✗
Reply-to message ID	✓	✗
Structural parser (not class-name fragile)	✓	✗
Price	PPE $3/1K messages	~$50/mo fixed

Frequently asked questions

Can I scrape a private Telegram channel?

No. This actor only works on public channels accessible via t.me/s/{channel}. Private channels, groups, and direct messages are not supported.

Does this require a Telegram bot token or phone number?

No. The actor scrapes the public Telegram web preview (t.me/s/*) — no account, no bot token, no API credentials needed.

How far back can I scrape?

As far back as Telegram's public web preview stores. Most active channels have several years of history available. Use maxMessages and beforeId cursor to control the range.

Does it download media files?

No. The actor detects media type (photo, video, document, sticker, audio, poll) and returns the type flag, but does not download or return media URLs. This keeps the output lean and the cost low.

How is pricing calculated?

Pay-per-result (PPE) at $3.00 per 1,000 messages ($0.003/message). One charge (message-scraped event) per message pushed to the dataset. Failed fetches, retries, and pagination probes are not charged.

Integrations

Built for brand monitors, crypto-signal aggregators, and LLM-pipeline builders ingesting public Telegram channel content — the JSON/dataset output drops into the tools you already run, no glue code:

• n8n / Make / Zapier — trigger a run or pipe every new dataset item into 500+ apps (Google Sheets, Airtable, Slack, HubSpot, your database) with no code: n8n, Make, Zapier.
• Webhooks — fire your own endpoint the moment a run finishes, to push results straight into your pipeline (docs).
• MCP server — expose this actor as a tool to Claude, Cursor, or any MCP client so an AI agent can pull this data mid-conversation (guide).
• API & SDKs — fetch the dataset as JSON, CSV, or Excel through the Apify REST API or the Python / JS SDKs.

See all Apify integrations.

Legal

This actor scrapes the publicly accessible web preview of Telegram channels (t.me/s/*). This is identical to what a browser sees when visiting the URL. Only public channels (already freely accessible to anyone on the internet) are scraped.

Not affiliated with Telegram.