Manage Kustomer customers, conversations, messages, notes, custom objects and CSAT surveys.
Kustomer ships in the w6w first-party pack. It declares 23 actions, 3 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.kustomerKustomer is a customer-service CRM built around the customer’s own timeline rather than a queue of tickets. This app covers that timeline end to end: creating and looking up customers, opening and updating conversations, recording messages on a conversation, adding internal notes, and tagging conversations for routing or reporting.
Beyond the conversation surface, it reaches Kustomer’s custom-object layer (Klasses and their KObject instances) for organizations that model domain data — orders, subscriptions, shipments — directly inside Kustomer, plus the access-management objects a workflow needs to route or attribute work: teams and users. CSAT survey configuration and individual satisfaction responses round out visibility into support quality.
Composing and sending outbound customer messages through a channel (email, chat, SMS, …) is deliberately out of scope: Kustomer models that through a separate, per-channel Draft resource with a materially different request shape per channel, which this app’s first version does not attempt.
Three routes to the same 23 actions. The Workflow tab is generated from Kustomer'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.
customer-find-by-external-id Look up a customer by the identifier your own system assigned it.
message-create Record a message on a conversation's timeline (a log entry — not composing/sending outbound content; see the app README).
satisfaction-response-get Fetch one customer's CSAT survey response by its Kustomer ID.
satisfaction-survey-list List the CSAT survey forms configured for your organization.
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 Kustomer's real ids.
{
"manifestVersion": "2",
"name": "kustomer-example",
"steps": [
{
"id": "conversation-add-tag",
"uses": {
"app": "io.w6w.kustomer",
"action": "conversation-add-tag",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"id": "<id>",
"tags": "<tags>"
}
}
]
}conversation-add-tag conversation-create conversation-get conversation-list conversation-update +18 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 Kustomer, 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 Kustomer. 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: "conversation-add-tag",
payload: {
id: "<value>",
tags: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action conversation-add-tag --payload '{"id":"<value>","tags":"<value>"}' Give an AI agent Kustomer — without giving it Kustomer'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.kustomer#conversation-add-tag",
"input": {
"id": "<id>",
"tags": "<tags>"
}
}
}
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 Kustomer 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.
Kustomer'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. Kustomer itself is MIT, and the runtime that executes it is source-available (FSL).
Kustomer declares its own checks, so its health is a property of the app rather than something the host guesses at.
Kustomer's own Statuspage.io incident history.
Per-organization allowance remaining, read off the `x-ratelimit-*` headers on the whoami probe.
Unauthenticated request to this connection's Kustomer org host. A 401 passes — it proves the org is serving; credential validity is the `auth:*` check's job.