Dust
List agents, run conversations, and search knowledge on Dust — the AI agent platform.
About
Dust ships in the w6w first-party pack. It declares 11 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
- App id
io.w6w.dust- Version
- 0.1.0
- Author
- w6w
- Licence
- MIT
- Categories
- AI & Machine Learning · Productivity
Overview
Dust is a workspace platform for building and running AI agents on your company’s own knowledge and tools. This app covers the agent/conversation surface a workflow needs: list, get and search agent configurations; create a conversation with an initial message (optionally mentioning an agent, and optionally waiting for its first answer); post follow-up messages into an existing conversation; cancel an in-progress generation; and attach inline text to a conversation as a content fragment for extra context.
It also covers read access to the workspace’s spaces (projects) and data sources, including semantic search over one data source’s documents — the same retrieval an agent’s own Search tool uses, callable directly from a workflow.
A Dust workspace lives in exactly one of two regions (US or EU), each with its own API host and its own key — this app’s connection form collects that alongside the API key and workspace id, and its health check separates a genuine outage from a region mismatch, which otherwise look identical.
Build with Dust
Three routes to the same 11 actions. The Workflow tab is generated from Dust'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.
Create Content Fragment
content-fragment-create Attach inline text (a document, transcript, or note) to a conversation as context.
Cancel Message Generation
conversation-cancel Stop one or more in-progress agent messages from continuing to generate.
Create Conversation
conversation-create Start a new conversation with an initial message, optionally mentioning an agent.
Get Conversation
conversation-get Retrieve a conversation and its messages, including agent replies.
List Data Sources
data-source-list List the data sources (folders, connections, websites) in a space.
Create Message
message-create Post a message into an existing conversation, optionally mentioning an agent.
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 Dust's real ids.
{
"manifestVersion": "2",
"name": "dust-example",
"steps": [
{
"id": "agent-get",
"uses": {
"app": "io.w6w.dust",
"action": "agent-get",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"sId": "<sId>"
}
}
]
}Here are some of the things you can do
-
Get Agent
readagent-get -
List Agents
readagent-list -
Search Agents
searchagent-search -
Create Content Fragment
performcontent-fragment-create -
Create Conversation
performconversation-create
+6 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 Dust, 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 Dust. 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: {
sId: "<value>",
// variant: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value);npm install -g @w6w/cliw6w run conn_YOUR_CONNECTION_ID --action agent-get --payload '{"sId":"<value>"}'Give an AI agent Dust — without giving it Dust'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.dust#agent-get",
"input": {
"sId": "<sId>"
}
}
}
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 Dust connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
What the agent gets
Credentials it can't read
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.
A tool surface scoped to the caller
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.
A durable workflow in one call
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.
Health-aware discovery
Dust'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. Dust itself is MIT, and the runtime that executes it is source-available (FSL).
Health checks
Dust declares its own checks, so its health is a property of the app rather than something the host guesses at.
Dust platform status
Component status from status.dust.tt, scoped to the REST API, Conversations and Data Sources components plus both regions. Excludes the deprecated Dust App Platform and the third-party connector/Slackbot/extension rows, which this app does not call.