Scheduling to Telegram through SkedCast's API#
SkedCast's /v1 REST API schedules text, photo, video, and document posts to Telegram the same way it does every other platform: one POST https://api.skedcast.com/v1/posts call with a content caption and a targets block naming the connected accounts.
You need a Telegram bot added as an admin of the channel, connected once through the console or through a connect link that POST /v1/clients/{id}/connect-invites creates.
The request is one contract for every connected platform. The fields that are specific to Telegram are listed below.
Example: schedule a post to Telegram#
{
"clientId": "00000000-0000-4000-8000-000000000001",
"content": "New build is out — changelog in the pinned message.",
"targets": { "mode": "accounts", "accountIds": ["00000000-0000-4000-8000-000000000002"] },
"schedule": { "type": "at", "scheduledAt": "2030-01-15T15:00:00Z" }
}Telegram facts#
| Fact | Value |
|---|---|
| Max caption length | 4,096 characters |
| Media per post | 10 media item per message. |
| Native platform scheduling | No. SkedCast holds the timer and publishes at the scheduled time. |
| SkedCast's default pacing | no daily cap, with at least 30 sec between posts, per account. This is SkedCast's per-account pacing, not Telegram's own published limit. See the rate limits reference. |
Published-post URL (permalink) | https://t.me/{channel}/{id} for a public channel and https://t.me/c/{internal}/{id} for a private channel or supergroup (it opens for members of that chat). A plain group or direct message has no link. |
Reading status and results#
GET https://api.skedcast.com/v1/post-targets/{id} returns the Telegram target's status (for example scheduled, on_hold, published, failed or missed). Once the post is published, externalId and permalink point at the live post (see the table above).
on_hold:holdReasonsays why (pacing, a platform limit, media still processing, your plan, or a channel that needs reconnecting) andretryAtsays when the next try is.missed: the scheduled time passed while the post waited.outcomeReasonsays why; reschedule it.failed:failureReasonsays why. When the post may already be live,outcomeReasonisdelivery_unconfirmed: check the account, because a retry is refused.requiredActionsays what you must do (reconnect_channel,change_plan,contact_support,reschedule,retryorcheck_platform), or isnullwhen nothing is needed.- Instead of polling, subscribe to the
target.publishedandtarget.failedwebhooks (one per destination;post.publishedandpost.failedreport the whole post). See the webhooks reference below. - Or connect an MCP client and ask an agent to check. See the MCP hub below.