Latest updates

Labs February 27, 2026

Postmark Skills: Teach AI coding agents how to send email

We released Postmark Skills — a set of open source Agent Skills that give AI coding agents accurate context about the Postmark API. Instead of guessing about endpoints and missing best practices, your agent gets production-ready knowledge about sending email, processing inbound messages, managing templates, and more.

Show details

Agent Skills are an open standard for teaching AI coding agents specialized tasks. They work with Claude Code, Cursor, GitHub Copilot, OpenAI Codex, Gemini CLI, and other tools that support the format.

Five skills are included:

postmark-send-email — Single, batch (up to 500), bulk, and template-based sending with proper Message Stream handling, error codes, and retry logic.
postmark-inbound — Reply-by-email, email-to-ticket, and document extraction workflows using inbound webhooks and MailboxHash routing.
postmark-templates — Server-side Handlebars templates with layout inheritance, validation, and cross-server pushing.
postmark-webhooks — Delivery, bounce, open, click, spam complaint, and subscription change event handling.
postmark-email-best-practices — Deliverability, compliance (CAN-SPAM, GDPR, CASL), email design, list management, and safe testing.

Get started:

npx skills add ActiveCampaign/postmark-skills

All five of our official SDKs are supported: Node.js/TypeScript, Python, Ruby, PHP, and .NET. Your agent detects your project language and uses the appropriate one.

This joins our MCP server, llms.txt file, and AI prompts as part of Postmark's AI tooling for developers. Where the MCP server lets AI tools interact with your Postmark account directly, skills teach AI agents how to write correct Postmark integration code.

MIT licensed and open to contributions on GitHub. Read the full blog post →