Postmark - Atako

Let your agents send transactional emails (plain or from a template), read message history, opens, bounces and suppressions, and follow delivery statistics of a Postmark server.

Connection

Authentication: API key (Server API token).

Sign in at account.postmarkapp.com → Servers → select your server → API Tokens → copy the Server API token. The token is tied to that server only (its messages, templates, bounces and stats). Sending requires a confirmed sender signature or domain on the account.See Postmark’s documentation.

Read actions (13)

Action	Description
`get_bounce`	Retrieve one bounce: type, email, details, dates, whether it can be reactivated (CanActivate).
`get_bounces_stats`	Bounce counts per day and per type (hard, soft, transient, SMTP API errors) for a period.
`get_message_opens`	List the open events of one sent message (date, client, OS, platform, geography). The full delivery timeline is in get_outbound_message (MessageEvents).
`get_opens_stats`	Open counts per day (total and unique) for a period.
`get_outbound_message`	Retrieve one sent message with its text and HTML bodies and its MessageEvents (Delivered, Transient, Opened, LinkClicked, Bounced, SubscriptionChanged).
`get_outbound_stats`	Outbound overview for a period: sent, bounced, bounce rate, spam complaints, tracked and unique opens, link clicks.
`get_server`	Retrieve the Postmark server this token belongs to (name, ID, tracking settings, inbound address). Cheapest call to check the token.
`get_template`	Retrieve one template (subject, HTML and text bodies, layout, alias) by numeric ID or alias.
`list_bounces`	List bounces (ID, type, email, date, whether it can be reactivated). type: e.g. HardBounce, SoftBounce, Transient, SpamComplaint, Blocked, AutoResponder. inactive: true for deactivated addresses only. email_filter, tag, message_id, from_date / to_date (Eastern Time, e.g. 2026-09-01T12:00:00), message_stream (default “outbound”).
`list_inbound_messages`	Search messages received by the server inbound address. Filters: recipient, from_email, tag, subject, mailbox_hash, status (blocked, processed, queued, failed, scheduled), from_date / to_date (e.g. 2026-09-01T12:00:00).
`list_outbound_messages`	Search sent messages (newest first): MessageID, recipients, subject, tag, status, stream, received date. Filters: recipient, from_email, tag, subject, status (queued, sent or processed), from_date / to_date (e.g. 2026-09-01T12:00:00), message_stream.
`list_suppressions`	List the suppressed recipients of a message stream (email, reason, origin, date). suppression_reason: HardBounce, SpamComplaint or ManualSuppression. origin: Recipient, Customer or Admin. from_date / to_date: YYYY-MM-DD.
`list_templates`	List the templates of the server (ID, alias, name, type, active). template_type: All (default), Standard or Layout. layout_template: keep only templates using this layout alias.

Write actions (3)

Action	Description
`reactivate_bounce`	Reactivate the address of a bounce so Postmark sends to it again. Only possible when the bounce has CanActivate: true (see get_bounce).
`send_email`	Send one transactional email. from: sender address with a confirmed sender signature (e.g. “Jane <jane@acme.com>”). to / cc / bcc: comma-separated addresses, at most 50 recipients across To, Cc and Bcc combined. subject: max 2000 characters. At least one of html_body or text_body is required. reply_to: comma-separated addresses. tag: max 1000 characters, used for statistics. track_opens: enable open tracking. message_stream: stream ID, default “outbound” (transactional). Attachments are not supported.
`send_email_with_template`	Send one email rendered from a Postmark template. Give exactly one of template_id (numeric) or template_alias. template_model: object of variables applied to the template to build its subject, HTML and text bodies. from: sender with a confirmed signature. to / cc / bcc: comma-separated addresses, max 50. message_stream: default “outbound”. Use list_templates to find IDs and aliases.

Permissions

Every action above must be explicitly granted to an agent before it can be used. See Permissions for the grant model and Security for how credentials are protected.

Last reviewed against the provider API: September 2026.