> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.beatapi.io/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.beatapi.io/_mcp/server.

# How should I supply Telegram channel usernames, links and message IDs?

> Pass plain channel usernames in channel and integer message IDs in post_id for Telegram list, detail, comment, and search actions, and store channel + post_id as the unique message key.

Pass a Telegram channel as the plain channel username, such as `example`, without `@` or a link prefix. Identify one message with the channel username plus the integer `post_id` from the message link: for [https://t.me/example/123](https://t.me/example/123), `channel` is `example` and `post_id` is `123`. `fetch_channel_info` alone also accepts `@durov` or a full channel URL.

## What belongs in `channel`?

`fetch_channel_info` accepts `durov`, `@durov`, and `https://t.me/durov`, normalizing all three inputs. For message lists, message details, comments and search, pass the plain channel username without `@`, without a URL prefix and without a message path. `telegram.web.batch_channel_info` takes `channels` as one comma-separated string of at most 20 usernames, not a JSON array.

## What identifies one message?

For [https://t.me/example/123](https://t.me/example/123), `channel` is `example` and `post_id` is the integer `123`. `telegram.web.fetch_post_detail` and `telegram.web.fetch_post_comments` both require `channel` and `post_id`. A message ID is unique only within one channel, so store and deduplicate messages by `channel` + `post_id`; message 123 in one channel and message 123 in another channel are different messages.

For the example link [https://t.me/example/123](https://t.me/example/123), the message-detail body is shown below. `example` illustrates the expected request format; replace `example` with the channel username from your permalink.

```json
{"action":"telegram.web.fetch_post_detail","params":{"channel":"example","post_id":123}}
```

To read the discussion, keep the same `channel` and `post_id` and call the comments action. The comments request is a separate paid call, not an extra field on the detail action:

```json
{"action":"telegram.web.fetch_post_comments","params":{"channel":"example","post_id":123,"limit":20}}
```

Keep the URL, the channel username and the integer message ID together in each source record. Apply the following identifier rules:

* A display name is not a username.
* The visible link of a forwarded message does not prove which channel published the original.
* A private invite link is not a channel username, and the numbers in an invite link grant no access.
* Do not add one to a message ID to fill a missing message. For a history task, use the cursors returned by the list action and set a separate request limit.

## Does retrieving a message also retrieve the message's comments?

No. `telegram.web.fetch_post_detail` does not call the comment action. Comments need a separate `telegram.web.fetch_post_comments` call, with `limit` 1–100 (default 50) and no `before` or `after` input. Do not pass message-list cursors to the comments action, and do not treat comment records as channel publications.

For example, an events channel forwards a product announcement. First save the forwarding channel and the forwarded message ID. To cite the publisher, open the original message the forward points to, and save the original channel and message ID as a separate record. When both records share the same number, the channel usernames tell the two messages apart.

[Pagination](/faq/telegram-pagination) · [Archive channel messages](/guides/telegram-channel-archive)