Work Dixa customer-support conversations, end users, agents, tags and queues.
Dixa ships in the w6w first-party pack. It declares 28 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.dixaDixa is a customer-service platform that routes email, chat, phone, SMS and messaging conversations to agents through queues. This app lets a workflow open an email conversation for an end user, read a conversation and its messages and internal notes, add a reply or a private note, assign it to an agent, move it to another queue, tag it, set its custom attributes, and close or reopen it.
It also manages the people on the other side: create, look up, list and update end users, list the conversations a given end user has had, and search conversations by text. Agents, tags, queues and custom-attribute definitions can be listed so a workflow can resolve names to the ids the other actions take.
Built on Dixa’s v1 REST API at dev.dixa.io with one API token.
Three routes to the same 28 actions. The Workflow tab is generated from Dixa'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.
agent-list List the agents and admins in the organisation. The email and phone filters are mutually exclusive — Dixa errors if both are given.
conversation-assign Assign (claim) a conversation to an agent. Without force, Dixa refuses to take it from an agent who already holds it.
conversation-create-email Open a new email conversation for an end user. An inbound message is authored by the requester; an outbound one by an agent.
conversation-custom-attributes-update Set custom attribute values on a conversation. Keys are custom-attribute definition ids (see Custom Attribute List); a value is a string, or an array of strings for a multi-select.
conversation-get Fetch one conversation: channel, state, requester, assignment, queue and custom attributes.
conversation-list-by-end-user List the conversations an end user has had, newest pages first as Dixa returns them.
conversation-search Full-text search over conversations. Each hit is `{ id, highlights }` — fetch the conversation with Get Conversation.
custom-attribute-list List custom attribute definitions (id, entityType, identifier, label, input definition). Their ids are the keys the custom-attribute update actions take.
end-user-custom-attributes-update Set custom attribute values on an end user. Keys are custom-attribute definition ids (see Custom Attribute List).
end-user-list List end users, optionally filtered by email, phone number or external id (the filters are alternatives — Dixa documents them for look-ups, not combined).
message-add Add a customer-visible message to a conversation. Outbound is an agent reply; inbound is posted as the end user.
message-list List a conversation's messages. Each carries a channel-specific `attributes` block (email, chat, phone, SMS…) discriminated by `_type`.
note-add Add a private internal note to a conversation. The end user never sees it.
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 Dixa's real ids.
{
"manifestVersion": "2",
"name": "dixa-example",
"steps": [
{
"id": "agent-get",
"uses": {
"app": "io.w6w.dixa",
"action": "agent-get",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"agentId": "<agentId>"
}
}
]
}agent-get agent-list conversation-assign conversation-create-email conversation-custom-attributes-update +23 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 Dixa, 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 Dixa. 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: "agent-get",
payload: {
agentId: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action agent-get --payload '{"agentId":"<value>"}' Give an AI agent Dixa — without giving it Dixa'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.dixa#agent-get",
"input": {
"agentId": "<agentId>"
}
}
}
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 Dixa 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.
Dixa'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. Dixa itself is MIT, and the runtime that executes it is source-available (FSL).
Dixa declares its own checks, so its health is a property of the app rather than something the host guesses at.
The `Dixa API and Exports API` component on Dixa's Statuspage. Every other component (channels, telephony, AI features) is reported as detail but never drives the verdict.