Mailchimp MCP Server

The most complete Model Context Protocol (MCP) server for the Mailchimp Marketing API: 227 tools to query and manage your Mailchimp account from any MCP-compatible client. It covers the full campaign, audience, member, segment, automation and reporting surface, complete e-commerce (read and write), landing pages, File Manager, surveys, signup forms and verified domains, plus multi-account support and runtime-security guardrails (read-only, dry-run and audit modes, with per-tool risk metadata).

Uses the Mailchimp Marketing API via requests. Not based on the official mailchimp-marketing-python client. I hit too many issues with it so I went with raw HTTP calls instead.

Quick start

uvx mailchimp-mcp   # no install required; just needs MAILCHIMP_API_KEY

Add it to your MCP client (Claude Desktop, Cursor, Cline, …):

{
  "mcpServers": {
    "mailchimp": {
      "command": "uvx",
      "args": ["mailchimp-mcp"],
      "env": {
        "MAILCHIMP_API_KEY": "your-key-us8",
        "MAILCHIMP_READ_ONLY": "true"
      }
    }
  }
}

Tip: start with MAILCHIMP_READ_ONLY=true to explore safely, then flip it off when you are ready to write. See Configuration for all options.

Docker

A prebuilt image is published to the GitHub Container Registry on every release:

docker run --rm -i -e MAILCHIMP_API_KEY=your-key-us8 ghcr.io/damientilman/mailchimp-mcp-server:latest

Or point your MCP client at it:

{
  "mcpServers": {
    "mailchimp": {
      "command": "docker",
      "args": ["run", "--rm", "-i", "-e", "MAILCHIMP_API_KEY", "-e", "MAILCHIMP_READ_ONLY", "ghcr.io/damientilman/mailchimp-mcp-server:latest"],
      "env": {
        "MAILCHIMP_API_KEY": "your-key-us8",
        "MAILCHIMP_READ_ONLY": "true"
      }
    }
  }
}

Features

Read

• Campaigns - List, search, and inspect campaign details
• Reports - Open/click rates, bounces, per-link clicks, domain performance, unsubscribe details
• Email activity - Per-recipient open/click tracking, member activity history
• Audiences - Browse audiences, members, segments, tags, and growth history
• Merge fields - List custom fields for an audience
• Interest categories & groups - Browse interest categories and options
• Webhooks - List configured webhooks
• Segments - Get segment details, conditions, and member lists
• Automations - List workflows, inspect emails in a workflow, view queues
• Templates - Browse templates and extract HTML content
• Landing pages - List and inspect landing pages
• File Manager - Browse stored images and files, and folders
• Surveys - List and inspect an audience's surveys
• Signup forms - Inspect an audience's signup forms
• E-commerce - Stores, orders, products, customers (requires e-commerce integration)
• Campaign folders - Browse folder organization
• Batch operations - Monitor bulk operation status

Write

• Members - Add, update, unsubscribe, delete, and tag contacts
• Audiences - Batch subscribe members, update audience settings
• Campaigns - Create drafts, set HTML content, schedule, unschedule, duplicate, delete, send, send test, cancel
• Segments/Tags - Create, update, delete, add/remove members, dynamic conditions
• Merge fields - Create, update, delete custom fields
• Interest categories - Create and delete categories and interests
• Webhooks - Create and delete webhooks
• Templates - Create, update, and delete email templates
• File Manager - Upload (base64) and delete images and files
• Surveys - Publish and unpublish audience surveys
• Signup forms - Customize header, contents, and styles
• Automations - Pause and start automation workflows
• E-commerce - Cart lifecycle, promo rules, and promo codes for discount workflows
• Batch - Run bulk API operations in a single request

Runtime security

• Risk metadata - Per-tool read / write / destructive classification via MCP annotations and describe_tools
• Audit log - Optional structured JSON audit events per dispatch (MAILCHIMP_AUDIT_LOG)
• Argument validation - Pagination caps and required-ID checks before every request

Prerequisites

• Python 3.10+
• A Mailchimp API key

Installation

Using uvx (recommended)

No installation needed — run directly:

uvx mailchimp-mcp

Using pip

pip install mailchimp-mcp

Then run:

mailchimp-mcp

From source

git clone https://github.com/damientilman/mailchimp-mcp-server.git
cd mailchimp-mcp-server
python -m venv .venv
source .venv/bin/activate
pip install -e .

Configuration

The datacenter (us8, us21, etc.) is automatically extracted from each key.

Safety modes

Read-only mode — When MAILCHIMP_READ_ONLY=true, all write tools (create, update, delete, schedule, etc.) are blocked and return an error. Read tools work normally. This is the recommended default for shared or exploratory setups where you only need reporting and analytics.

Dry-run mode — When MAILCHIMP_DRY_RUN=true, write tools return a preview of the action they would perform (tool name, target resource, parameters, and its risk tier) without making any API call. Useful for testing prompts before going live.

Runtime security

The server is designed for defense-in-depth alongside an external MCP gateway: it owns the semantics (which tools are reads, reversible writes, or irreversible) and exposes them as machine-readable signals a gateway can enforce on, rather than making the gateway guess which calls are dangerous.

Risk metadata — Every tool is classified as read, write, or destructive (irreversible data loss or an irreversible send). This surfaces two ways:

• MCP tool annotations — readOnlyHint, destructiveHint, and idempotentHint travel through the standard tools/list, so any compliant client or gateway reads them directly.
• describe_tools — a read tool returning {name, risk, read_only, destructive, idempotent} for every tool plus per-tier counts, for gateways that prefer a tool call.

Structured audit log — With MAILCHIMP_AUDIT_LOG=true, each dispatch emits one JSON event to stderr with the tool, its risk tier, the destructive flag, the target account, the outcome (executed / blocked_read_only / dry_run), and the inspected arguments. Bulky or sensitive values (e.g. base64 file_data) are redacted and response bodies are never logged, so the stream is a safe audit sink to tail centrally.

Argument-contract validation — All writes funnel through a single _guard_write chokepoint, and every request is validated before dispatch (pagination count capped to 1–1000, missing path IDs rejected), so malformed calls fail fast and consistently instead of hitting the API.

Tool profiles

Exposing all 227 tools sends a large tool list to the model on every turn (~63k tokens). If you only need part of the surface, load a subset with one line:

"env": { "MAILCHIMP_API_KEY": "your-key-us8", "MAILCHIMP_TOOLS": "read" }