First-party app
JumpCloud

JumpCloud

Manage JumpCloud users, devices, groups and remote commands from one directory.

stable Security & IdentityDevOps & InfrastructureHuman Resources

About

JumpCloud ships in the w6w first-party pack. It declares 31 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.

App id
io.w6w.jumpcloud
Version
0.1.1
Author
w6w
Licence
MIT
Categories
Security & Identity · DevOps & Infrastructure · Human Resources

Overview

JumpCloud is a cloud directory, and this app manages the users, devices, groups, and remote commands that make it one: creating, updating, and offboarding users; enrolling and inspecting devices; running saved commands against them; and managing the static and dynamic groups that bind users and systems together. It is built for identity and device-lifecycle workflows — provisioning a new hire, suspending someone who’s left, or pushing a command out to a fleet of machines — across whichever regional console an organization’s directory lives in.

A few actions are deliberately separated from their safer neighbors because of what they can undo: wiping a device, permanently deleting a user, and deleting a group that carries every application and device binding it granted are each their own action requiring explicit confirmation, distinct from suspending a user or simply unenrolling a device. Command results are read separately from running them, since a queued command against an offline device can report success without having done anything yet.

Build with JumpCloud

Three routes to the same 31 actions. The Workflow tab is generated from JumpCloud'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.

Get a command

command-get

Retrieve a saved command, its script and the devices it is bound to.

List commands

command-list

List saved commands, which are scripts with a target set.

List command results

command-result-list

List the outcomes devices reported for commands, with exit codes and output.

Run a command

command-run

Queue a saved command against named devices, or its own bound targets.

List organizations

organization-list

List the organizations this key can act on — the ids the connection's org field takes.

Lock, restart or shut down a device

system-command

Send a builtin device command. Queues if the device is offline.

Delete a device

system-delete

Unenrol a device and uninstall its agent. Does not wipe the machine.

Erase a device

system-erase

Remotely wipe a device. Queues and fires later if the device is offline.

Get a device

system-get

Retrieve one enrolled device by id.

List device groups

system-group-list

List device groups, which policies and commands attach to.

Add or remove a device from a group

system-group-member-set

Move a device in or out of a group, applying or removing that group's policies.

List devices

system-list

List enrolled devices, optionally filtered.

Update a device

system-update

Rename a device, retag it, or change its SSH login policy.

Create a user

user-create

Create a directory user, staged by default.

Delete a user

user-delete

Permanently delete a directory user. Suspending is usually what offboarding wants.

Get a user

user-get

Retrieve one directory user by id.

Create a user group

user-group-create

Create a static user group, or a dynamic one driven by an attribute rule.

Delete a user group

user-group-delete

Delete a user group and every access binding it carried.

Get a user group

user-group-get

Retrieve one user group, including whether its membership is dynamic.

List user groups

user-group-list

List user groups. Membership in one is how access is granted.

List a user group's members

user-group-member-list

List a group's direct members, or everyone effectively in it.

Add or remove a group member

user-group-member-set

Grant or revoke access by changing a user's membership of a group.

List users

user-list

List directory users, optionally filtered.

Reset a user's MFA

user-mfa-reset

Clear the enrolled TOTP factor so the user can enrol a new authenticator.

Expire a user's password

user-password-expire

Force a password change at next login. Does not end current sessions.

Set a user's password

user-password-set

Set a password directly. The user is not emailed.

Add an SSH key to a user

user-sshkey-add

Push a public SSH key to every device this user is bound to.

List a user's SSH keys

user-sshkey-list

List the public SSH keys JumpCloud pushes to this user's devices.

Activate or suspend a user

user-state-set

Turn a user's access on or off without deleting the account.

Unlock a user

user-unlock

Clear a lockout caused by failed login attempts. Does not lift a suspension.

Update a user

user-update

Change a user's profile fields. Does not change their state.

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 JumpCloud's real ids.

{
  "manifestVersion": "2",
  "name": "jumpcloud-example",
  "steps": [
    {
      "id": "command-get",
      "uses": {
        "app": "io.w6w.jumpcloud",
        "action": "command-get",
        "connection": "conn_YOUR_CONNECTION_ID"
      },
      "with": {
        "commandId": "<commandId>"
      }
    }
  ]
}

Here are some of the things you can do

  • Get a command

    read
    command-get
  • List commands

    read
    command-list
  • List command results

    read
    command-result-list
  • Run a command

    perform
    command-run
  • List organizations

    read
    organization-list

+26 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 JumpCloud, 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 JumpCloud. The Workflow tab is where this app's real ids are.

Install
npm install @w6w/sdk
yarn add @w6w/sdk
pnpm add @w6w/sdk
deno add npm:@w6w/sdk
Code
import { 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: "command-get",
  payload: {
    commandId: "<value>",
  },
});

if (isActionRun(envelope)) console.log(envelope.value);
Install the CLI
npm install -g @w6w/cli
CLI
w6w run conn_YOUR_CONNECTION_ID --action command-get --payload '{"commandId":"<value>"}'

Give an AI agent JumpCloud — without giving it JumpCloud'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.jumpcloud#command-get",
    "input": {
      "commandId": "<commandId>"
    }
  }
}

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 JumpCloud 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

JumpCloud'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. JumpCloud itself is MIT, and the runtime that executes it is source-available (FSL).

Request MCP access

Health checks

JumpCloud declares its own checks, so its health is a property of the app rather than something the host guesses at.

service

JumpCloud platform status

The API, command and group components for THIS connection's region on JumpCloud's status page. Reads the Connection for the region; sends no credential.

quota

API request headroom