Run Hex projects, poll or cancel runs, and read projects, users, groups, collections and data connections.
Hex ships in the w6w first-party pack. It declares 15 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.hexHex is a collaborative analytics workspace where notebooks and data apps run SQL and Python against a team’s warehouse. This app lets a workflow trigger a published Hex project on demand, hand it input parameters, poll the run until it reaches a terminal state, and cancel it if it should not finish.
Around that core it reads the workspace itself: projects and their metadata, the runs of each project filtered by status or trigger, the warehouse tables a project queries, and the users, groups, collections and data connections that organise access.
A run always executes the latest published version of a project and answers with a run handle rather than a result, so a workflow starts a run, polls its status, and reads results where the project publishes them. Health checks track Hex’s status page, reading the components the API and runs depend on.
Three routes to the same 15 actions. The Workflow tab is generated from Hex'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.
me-get Fetch the user the token belongs to, the workspace id and the token's expiry. A workspace token returns no user fields.
project-get Fetch one project's metadata: title, owner, status, categories and view analytics.
project-list List Hex projects the token can view, with optional status, owner and collection filters.
project-queried-tables-list List the data-warehouse tables a project queries.
project-run Trigger a run of a project's latest published version and return the run handle immediately.
project-runs-list List runs of a project, optionally filtered by status and trigger type.
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 Hex's real ids.
{
"manifestVersion": "2",
"name": "hex-example",
"steps": [
{
"id": "collection-get",
"uses": {
"app": "io.w6w.hex",
"action": "collection-get",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"collectionId": "<collectionId>"
}
}
]
}collection-get collection-list data-connection-get data-connection-list group-get +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 Hex, 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 Hex. 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: "collection-get",
payload: {
collectionId: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action collection-get --payload '{"collectionId":"<value>"}' Give an AI agent Hex — without giving it Hex'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.hex#collection-get",
"input": {
"collectionId": "<collectionId>"
}
}
}
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 Hex 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.
Hex'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. Hex itself is MIT, and the runtime that executes it is source-available (FSL).
Hex declares its own checks, so its health is a property of the app rather than something the host guesses at.
Component status from status.hex.tech. Main site (hosts the API), Kernels and Data Connections decide the verdict; Login and Single tenant stacks are shown as detail.