Convoy AI

Process Station · HR

Employee onboarding

CommunityBuilt with LauncherTemplate

Accounts, access and equipment, in order, with manager approval.

The control station, in brief

This station links several tools into one journey. The agent follows the steps in order, and every step goes through the rules below.

Linked toolsDirectoryMailEquipment

It can

  • Create the account
  • Order equipment

It asks a human for

  • Grant access

It never has access to

  • Grant access before the account exists

Scenario Replays

An agent works through the station. Simulated scenarios, no real data, no key required.

Onboard a new hire

What you will see: Three tools, one journey: the order is enforced and the manager keeps the final key.

Agent connected to the station through MCP
  1. HumanPrepare Léa's arrival on Monday, developer.
  2. AgentI'll start with the account, then equipment, then access.
  3. MCP callcreateAccount(name: "Léa", role: "dev")
  4. StationAccount created in the directory.OKRule applied: playbook onboard-an-employee
  5. MCP callorderEquipment(kit: "dev")
  6. StationLaptop and screen ordered.OKRule applied: playbook onboard-an-employee
  7. MCP callgrantAccess(account: "lea", apps: 4)
  8. StationAccess is ready. It opens after the manager approves, who has received the request.PENDINGRule applied: manager-approval
These rules are not yours?

Copy the station, change a limit or an approval step in its configuration file, and get your own station.

Copy and adapt

The configuration in detail

employee-onboarding.yamlRead the docs →

Everything the station does fits in this file. The replay above applies it, and a copy starts from it. The values are examples to adapt, not a recommendation.

station:
  id: employee-onboarding
  kind: process
  policies:
    - {policy: manager-approval, at: submit}
    - {policy: account-first, at: submit}

global:
  language: en
  timezone: Europe/Paris

apis:
  - id: directory
    operations:
      createAccount:       {access: write, risk: medium}
      grantAccess:         {access: write, risk: high}
  - id: equipment
    operations:
      orderEquipment:      {access: write, risk: medium}

profiles:
  hr-assistant:
    allow: [createAccount, grantAccess, orderEquipment]

policies:
  manager-approval:
    type: rule
    when: "action == grantAccess"
    effect: needs-human
    reason: "Access is granted after the manager's approval."
  account-first:
    type: rule
    when: "action == grantAccess and not account_exists"
    effect: deny
    reason: "Access is granted only on an existing account."

skills:
  - {id: create-account, uses: createAccount}
  - {id: grant-access, uses: grantAccess}
  - {id: order-equipment, uses: orderEquipment}

playbooks:
  - id: onboard-an-employee
    tasks: [create-account, order-equipment, grant-access]

Shared copies

Copies that their authors choose to share will appear here. Each one keeps a link to this original station.