Any RSS feed, posted to a Telegram chat or channel.
You want new releases, advisories or blog posts to show up in the Telegram group or channel you already read. A hand-rolled poller means a seen-items table, a cron entry and the bug where it re-posts everything after a deploy. Point emit at the feed and give it your bot's sendMessage URL instead.
One call
curl -X POST https://api.rssemit.com/v1/sinks \
-H "Authorization: Bearer $EMIT_KEY" \
-H "Content-Type: application/json" \
-d '{
"kind": "telegram",
"url": "https://api.telegram.org/bot123456789:XXXX/sendMessage?chat_id=-1001234567890",
"feed_url": "https://github.com/nodejs/node/releases.atom"
}'
The feed here is a Node.js releases feed from the developer feed catalog, but any RSS, Atom or JSON Feed URL works — including feeds you don't own, and feeds emit built for you from a page that never had one.
emit posts each item as a plain-text message with the item title, link and a short summary, capped at 1,900 characters, with link previews disabled. Templates work as for the other chat destinations.
The URL is a secret. The bot token is part of the URL path and authorizes your bot. Error text emit stores, logs or returns is redacted, and the full URL is returned only to the authenticated owner of the sink.
Set up the Telegram side
-
Create a bot with BotFather and copy its token (it
looks like
123456789:AA…). -
Add the bot to your group or channel with permission to post, then find the chat: a
numeric ID (group IDs are negative) or an
@channelusername. -
Build the destination
urlashttps://api.telegram.org/bot<token>/sendMessage?chat_id=<id>. It must be HTTPS onapi.telegram.orgwith a bot-token path and exactly onechat_id; anything else is rejected with a 422. -
Send a test with
POST /v1/sinks/{id}/testto confirm the message lands in the chat before you wait for the next item.
Or one digest a day
A busy feed doesn't need to ping the channel forty times. Add a schedule and emit holds new items and posts one message at the boundary:
curl -X POST https://api.rssemit.com/v1/sinks \
-H "Authorization: Bearer $EMIT_KEY" \
-H "Content-Type: application/json" \
-d '{
"kind": "telegram",
"url": "https://api.telegram.org/bot123456789:XXXX/sendMessage?chat_id=-1001234567890",
"feed_url": "https://github.com/nodejs/node/releases.atom",
"schedule": "daily",
"digest_hour": 9,
"digest_tz": "America/New_York"
}'
Use "schedule": "weekly" with digest_weekday (0 is Monday) for a
weekly roundup. Without a schedule, each item is posted as it arrives.
The parts you were going to get wrong
- No back-catalog blast. The first poll records a silent baseline, so connecting a feed with ten years of history posts nothing.
- Each item once. Items are deduplicated by guid, so republished or reordered entries don't repeat.
- Failures retry. If Telegram is down or rejects the post, the item stays queued and is retried on the next cycle. The dashboard shows the last error.
- Gentle on the publisher. Feeds are polled every five minutes with conditional GET, so a quiet feed costs a 304, not a download.
Pricing
250 sink deliveries a month free. Polling feeds costs nothing, and each account gets 250 free sink deliveries per calendar month (UTC) across Slack, Discord, Teams, Telegram, Google Chat and JSON sinks. After that each delivery is 1 credit ($1.60 per 1,000 credits); a daily or weekly digest counts as one delivery however many items it carries, test sends are free, and failed posts are not charged. If the allowance is used and the balance can't cover more, deliveries pause and items are held, not dropped, until you add credits.
This is a chat sink, not a webhook of yours
A Telegram sink posts a message through your bot to a chat. It is separate from the JSON sink, which POSTs a feed's items to your own endpoint, and from the signed account-event webhooks in the API guide, which report things that happened in your emit account.
Other chat destinations
The same call works for other chat apps: change kind and the destination
url. Slack, Discord, Teams and Google Chat all work this way. Setup for all of
them is in the API guide, and they share the same allowance,
digests and retries.