When I Work

Connect AI agents to When I Work employee scheduling. Agents manage users, create and review shifts, pull time entries, and look up locations and positions to keep hourly teams staffed.

9 actions API key available

When I Work website ↗ Platform Documentation ↗ Capability data verified 2026-07-28

A new hire is added in the HR system with a start date and role

List Users confirms no existing account under the same email

List Locations and List Positions resolve the hire's worksite and job role IDs

Create User adds the employee, which sends their activation invitation

Create Shift drafts their first week of shifts, unpublished

The scheduling manager reviews the drafted shifts against the location's coverage and labor budget before anything is assigned or published

A welcome email goes out with start details, and the schedule posts to the team channel

What This Integration Enables

Hourly scheduling is where staffing math meets people's actual lives, and When I Work is the system that holds both. FlowRunner agents handle the mechanical half of that job: they create users when the HR system hires, build and review shifts against real location and position data, pull clock in and out records for payroll and attendance, and keep the schedule flowing out to the channels the team actually reads. What agents do not do is decide who works when, because that decision carries context no API returns: who asked for Saturday off, who is close to overtime, who covered last weekend. - Onboard new hires into scheduling automatically when [BambooHR](/integrations/bamboohr) creates them - Draft first-week shifts as unpublished proposals for a manager to approve - Export a week's schedule for a location or a single employee to [Slack](/integrations/slack) or a report - Pull timesheet entries into [Google Sheets](/integrations/google-sheets) for payroll without manual exports - Power employee pickers and lookups in downstream automations with live user data The unpublished-by-default behavior of new shifts is the connector's most important property, because it lets agents do the drafting at full speed while keeping publication, the moment a schedule becomes a promise, in a manager's hands.

Without FlowRunner

Scheduling in two systems The hire exists in HR but not in scheduling, and the gap is bridged by retyping

Payroll prep is a copy job Clock in and out records get exported, cleaned, and pasted into the payroll sheet by hand

The schedule lives in one app Employees who missed the app notification find out about shifts from a coworker

With FlowRunner

One record, both systems The HR event creates the scheduling account and drafts the first shifts automatically

Timesheets flow to payroll Time entries land in the payroll sheet on a schedule, bounded, filtered, and complete

The schedule finds the team Published shifts post to Slack and email, so the schedule reaches people where they already are

Use Case Scenarios

Hired Tuesday, scheduled by Wednesday

[BambooHR](/integrations/bamboohr) fires Create Employee for a new cashier starting next Monday. The agent checks List Users for duplicates, resolves the store via List Locations and the role via List Positions, and calls Create User, which sends the activation invite. It then drafts the first week with Create Shift, unpublished, mirroring the store's standard pattern for that position. The scheduling manager gets the draft, adjusts Thursday, and publishes. [Gmail](/integrations/gmail) sends the welcome with start details attached.

Payroll Friday without the export

Every Friday morning, List Times pulls the pay period's clock in and out entries for each location, bounded by the period's start and end. Each entry lands as a row in [Google Sheets](/integrations/google-sheets): user, shift, length, location, position. Entries with no associated shift, or lengths that diverge from the scheduled shift by more than the grace window, get flagged in a separate tab for the manager to resolve before payroll runs. The payroll clerk starts from a clean sheet instead of a raw export.

The schedule that posts itself

Sunday evening, List Shifts pulls the coming week for each location. The agent formats it by day and position and posts it to the location's [Slack](/integrations/slack) channel, with open shifts called out explicitly so anyone wanting extra hours can claim them with the manager. Employees stop learning their schedule from screenshots of screenshots.

Human-in-Loop Highlight

Create Shift with a User ID attached puts hours on a specific person's calendar and, once the schedule is published, a commitment on their phone: be at this location, in this position, at this time. That write spends two things at once, the labor budget and an employee's plans, and unwinding it after publication means a manager apologizing to someone who already arranged childcare. So FlowRunner flows exploit a property the platform itself provides: new shifts are drafts until published. Agents create shifts unassigned or unpublished, then hand the manager a coverage view, "Monday needs two servers, the draft assigns Kim (32 hours so far) and Deshawn (38 hours, one more shift hits overtime)," and the manager assigns, swaps, and publishes. The agent does the pattern-matching and the arithmetic. The person makes the promise, because the promise is to a person.

Agent processes routinely

Detects exception requiring judgment

Clear match Continues automatically

Ambiguous Routes to human via preferred channel

Human decides

Agent resumes with decision

Agent Capabilities

9 actions

Users

List Users Retrieves employees with filters for location, role, and account status, returning each user's ID, name, email, role, and status. The duplicate check before onboarding and the roster behind every sweep.
Get User Retrieves one employee in full by ID: name, email, role, phone, employee code, and account status. The context read before edits, escalations, and welcome messages.
Create User Creates a new employee, who typically receives an activation invitation. Requires email, first name, last name, and role. The onboarding write that bridges the HR system and the schedule.

Shifts

List Shifts Retrieves scheduled shifts in a date window, filterable by user, location, and position. The read behind schedule exports, coverage checks, and team channel posts.
Get Shift Retrieves one shift in full by ID: assigned user, location, position, times, break, notes, and publish status. Used to verify exactly what an employee was scheduled for before comparing against time entries.
Create Shift Creates a scheduled shift, assigned to a user or left open, at a location and position with start and end times. New shifts start unpublished, which is precisely the property FlowRunner's draft-then-approve pattern is built on.

Time and Attendance

List Times Retrieves clock in and out records in a date window, filterable by user and location, with each entry's length and associated shift. The payroll and attendance feed.

Account Structure

List Locations Retrieves the account's worksites with ID, name, address, and timezone. Reference data for creating shifts and scoping reports.
List Positions Retrieves the account's job roles, such as Cashier or Server, with ID, name, and color. Reference data for shift creation and schedule formatting.

Frequently Asked Questions

What can FlowRunner do with When I Work?

FlowRunner agents can run List Users, Get User, and Create User in When I Work, plus 6 more actions.

Does connecting When I Work to FlowRunner require OAuth?

No. When I Work connects to FlowRunner with an API key, no OAuth flow required.

Can When I Work trigger a FlowRunner workflow automatically?

When I Work doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.

Start building with When I Work

$100 in credits. No card required. Connect in minutes.

Start Building Free Talk to us