Manage Mailchimp audiences: create/update members, tag them, and send campaigns.
Mailchimp ships in the w6w first-party pack. It declares 15 actions, 3 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.mailchimpMailchimp manages an audience and its campaigns from a workflow — creating and updating members, tagging them (and untagging them) to drive segmentation, and reading a member’s or a campaign’s current state without opening Mailchimp’s dashboard.
Sending is covered from more than one angle: a campaign can be sent outright, resent to non-openers, or replicated into a new draft first — the pattern for reusing a campaign’s design without editing the original. Group interests can be listed to see how an audience is segmented before targeting a send at part of it.
A fit for growing and tagging an audience as people sign up elsewhere, triggering a campaign send from an external event, or keeping member data in sync between Mailchimp and another system.
Three routes to the same 15 actions. The Workflow tab is generated from Mailchimp's own manifest and carries its real ids, so it is copy-pasteable; the Code and CLI examples are the same call for any action on any app, so every app-specific value in them is a blank you fill in.
resend-campaign Create a resend-to-non-openers copy of a sent campaign.
A workflow step names the app and the action, and the editor fills in the
connection when you pick one. This is the Step shape from the
workflow spec, carrying Mailchimp's real ids.
{
"manifestVersion": "2",
"name": "mailchimp-example",
"steps": [
{
"id": "create-member",
"uses": {
"app": "io.w6w.mailchimp",
"action": "create-member",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"listId": "<listId>",
"email": "<email>",
"status": "<status>"
}
}
]
}create-member create-member-tag get-campaign get-member list-campaigns +10 more actions available
Every app-specific value here is a blank you have to fill in. An
app action is reached through the connection that authenticates it, so the
address is a connection id, not the app id — and connections belong to your account,
so a public page cannot know yours. Create one for Mailchimp, then fill in
the three blanks: conn_YOUR_CONNECTION_ID, the action key, and the
parameters that action declares. The call itself is real — the shape is transcribed
from the studio's own snippet builder, which prints the same kind of blanks — but
nothing in it is specific to Mailchimp. The Workflow tab is where this app's
real ids are.
npm install @w6w/sdkyarn add @w6w/sdkpnpm add @w6w/sdkdeno add npm:@w6w/sdkimport { W6wClient, isActionRun } from "@w6w/sdk";
// Reads W6W_BASE_URL and W6W_TOKEN from the environment when omitted.
const client = new W6wClient();
const envelope = await client.run({
urn: "conn_YOUR_CONNECTION_ID",
action: "create-member",
payload: {
listId: "<value>",
email: "<value>",
status: "<value>",
// emailType: "<value>",
// language: "<value>",
// vip: "<value>",
// ipSignup: "<value>",
// ipOptIn: "<value>",
// timestampSignup: "<value>",
// timestampOpt: "<value>",
// tags: "<value>",
// location: "<value>",
// mergeFields: "<value>",
// interests: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action create-member --payload '{"listId":"<value>","email":"<value>","status":"<value>"}' Give an AI agent Mailchimp — without giving it Mailchimp's credentials. One MCP endpoint exposes every app, function and workflow the caller is entitled to, as tools it can discover and run. Access is granted per team while we onboard.
One tool call{
"name": "w6w_invoke",
"arguments": {
"ref": "app:io.w6w.mailchimp#create-member",
"input": {
"listId": "<listId>",
"email": "<email>",
"status": "<status>"
}
}
}
Every tool names its target with a single ref. The
app: form above doesn't name a connection at all — the
host resolves which of the caller's Mailchimp connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
The token is attached host-side, at the moment of the call. It is never a tool argument, never in the model's context, and never in a transcript — so a prompt injection has nothing to exfiltrate.
Tools are derived per end user from what that person has actually connected and is entitled to — not one shared bot identity carrying the union of everyone's access.
Multi-step work runs on the workflow engine and returns a run handle the agent can poll — retries, branching and state survive the conversation that started them.
Mailchimp's declared health checks are on the surface too, so an agent can tell "the vendor is down" from "your credential expired" before it burns a retry on either.
The MCP surface is part of the hosted platform. Mailchimp itself is MIT, and the runtime that executes it is source-available (FSL).
Mailchimp declares its own checks, so its health is a property of the app rather than something the host guesses at.