Telegram Public Channel History & Watchlist

Export a watchlist of public broadcast-channel posts for news monitoring and competitor content research. One message row contains privacy-screened text, source date/link, approximate views, public reply context, visible reactions and media-type flags. This unofficial tool reads anonymous t.me/s/ web previews and follows their older-page links.

Input

Provide 1–5 channels as usernames, @handles or public t.me/s/CHANNEL URLs. Empty input alone defaults to channels=["telegram"]. maxMessagesPerChannel defaults 100, maxItems 500. Set UTC ISO 8601 since/until, or pass afterMessageIds keyed by normalized channel for incremental monitoring. keywords filters text within known channels; this does not search all Telegram or discover channels. Public groups, private/invite links, phone URLs and accounts are rejected. Order is descending public channel post ID, not inferred edit chronology.

Output

Each unique message has channel, messageId, sourceUrl, redacted text/flag, publishedAt, approximate views and viewsRaw, edited marker, same-channel replyToMessageId if visible, aggregate reactionCounts, mediaTypes, sanitized externalUrls, public_web_preview coverage, sourceRank and observedAt. Optional unavailable fields are omitted. Album media with one post ID is one row/event. No author/signature, forwarded person, subscriber identities, private replies or media files are returned. Uncharged SUMMARY records per-channel pages/counts, skipped/redacted counts, first/last message IDs, next older boundary and cap/source/date/checkpoint stop. Advance the checkpoint to the highest delivered message ID only after the intended interval is fully reconciled. If SUMMARY reports source_cap, older messages in that interval may remain undelivered: keep the previous checkpoint until they have been recovered and checked. Run often enough or reconcile the interval separately before advancing; a capped export alone is not a complete checkpoint.

Charging

$0.0003 per distinct delivered message through message-exported; $0.30/1000. No actor-specific start event or trial. Collect all requested sources into a bounded sanitized buffer before output/billing. Required-source failures leave zero rows/events. Duplicates, filters, metadata and SUMMARY incur no event. If no deliverable messages across all channels, fail without charging. Mixed genuine empty windows are recorded uncharged when other channels produce rows.

Limits

256MB, HTTP/HTML, no login, bot token, API credentials, paid API, proxy, browser or downloads. About 20 visible messages/page; follow actual before links with strict decreasing-ID/no-repeat checks.80-request budget,1-second pacing,20-second request timeout,2MiB response cap and 300-second run timeout. Per-channel 200/global 1000-row ceiling. Real source/date/checkpoint boundaries can deliver fewer than the ceiling. Logs contain counts/status only. Missing identity/content, login redirect,429, changed markup and nonprogress fail explicitly. No cross-run checkpoint storage.

Privacy

Exclude sender names/IDs/signatures, forwarded-person identities, member/subscriber identifiers, profile bios/avatars, contact addresses, phones, credentials and precise location. Requested broadcast-channel handle and public message link are the only identity references for provenance. Redact email/phone/@mention/contactURL/secret-like substrings before persistence. Discard raw response bodies and exclude source-restricted preview records. Pattern checks cannot guarantee prose names never appear; this is contact minimization, not anonymization or identity research.

Source and terms

Reads only public t.me/s/ broadcast previews and older before links. t.me robots currently 404; check every run and honor future Disallow. Telegram content-licensing terms restrict user-content scraping, aggregation and AI uses. The scout records the exact applicable excerpt and DR-0007 treatment. No login/closed-chat/paywall bypass or media download. This is current owner-selected channel monitoring, not an AI training or platform-wide content archive product. An AI agent can retrieve announcements for a requested channel comparison using the same limited rows. DC is unverified; independent review, private default/nondefault runs and PPE/dataset parity remain release gates.

If something goes wrong

• Wrong number or a failed run? Open a ticket on the Issues tab. I read every one and reply within 2 business days (Japan time).
• You never get a fake "empty" result. If the site can't be read, the run fails and says so.
• Delivered rows only. One validated unique row is one billable event. Duplicates, summaries, failed acquisition and undelivered rows are not charged.
• Checked every week. An automatic test runs this tool weekly; if the site changes, I fix it.
• Public source data only. No login or member/applicant records. Contact patterns are screened before delivery; prose may still name people.

More tools by the same author

• Mercari Japan Sold Prices — What Items Really Sell For
• Yahoo! Auctions Japan Sold Prices — Median, Range, Bids

All tools (Japan marketplaces, real estate, jobs, racing, prediction markets): https://apify.com/jpmarketdata

Disclaimer

Unofficial, independent tool — not affiliated with, endorsed by, or sponsored by Telegram. Product names and logos belong to their owners and only say where the data comes from. Data is read from public pages, for market research; check before you act on it.