Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@EmailOctopus MCP Server How did last Tuesday's newsletter do? Which links got clicked?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A Model Context Protocol server for EmailOctopus. It lets Claude, Cursor, ChatGPT and other AI agents manage your lists and contacts, read campaign results, and start automations.
Unofficial. This is a community project and is not affiliated with EmailOctopus. It was built from the public API v2 docs.
"How did last Tuesday's newsletter do compared to the one before? Which links got clicked?"
"Tag everyone who clicked the pricing link in campaign X as hot-lead."
"Add these 12 webinar attendees to the Newsletter list with their first names and tag them webinar-oct."
"Who signed up this week? Summarize by source field."
"Unsubscribe jane@acme.com from all lists."
Related MCP server: Sequenzy MCP Server
Tool | What it does | Writes? |
| Lists with counts, fields and tags | No |
| Create a mailing list | Yes |
| Segment by tag, status, created/updated date range; paginated | No |
| Look up by email address or ID | No |
| Add or update by email, with fields, tags to add/remove, and status | Yes |
| Change email, fields, tags or status (for example, unsubscribe) | Yes |
| Permanently delete a contact | Destructive |
| Campaign list and details | No |
| Summary stats plus per-link clicks in one call | No |
| Who opened / clicked / bounced / did not open, etc. | No |
| Queue a contact into an API-triggered automation | Yes |
Notes for agents and humans:
Contacts can be referenced by email address. The server converts emails to the MD5 form EmailOctopus expects.
Tags are sent as simple add_tags / remove_tags arrays. The server builds EmailOctopus' {tag: true/false} patch format for you.
Destructive tools carry MCP destructiveHint annotations, so clients can ask before running them.
Create an API key at EmailOctopus → Settings → API keys.
Build it:
git clone https://github.com/OfirOhan/emailoctopus-mcp.git
cd emailoctopus-mcp && npm install && npm run build{
"mcpServers": {
"emailoctopus": {
"command": "node",
"args": ["/absolute/path/to/emailoctopus-mcp/dist/index.js"],
"env": { "EMAILOCTOPUS_API_KEY": "your_api_key" }
}
}
}claude mcp add emailoctopus -e EMAILOCTOPUS_API_KEY=your_api_key -- node /path/to/emailoctopus-mcp/dist/index.jsnpm install
npm test # builds, runs unit tests and an end-to-end MCP stdio test against a fake EmailOctopus APIThe tests run on Node 20, 22 and 24 in CI.
Built by Ofir Ohana, an AI agents engineer. Issues and PRs are welcome.
MIT
This server cannot be deployed
AI-controlled email ops for campaigns, contacts, segments, analytics, and sending.
Email, WhatsApp and Telegram for AI agents: send, campaigns, automations, contacts, agent inboxes.
Mailchimp engagement intelligence for AI agents: campaign analytics and audience insights.
Read audiences, members, campaigns and reports; add, update, tag and archive subscribers.
