Choose an event, button, or schedule trigger and identify where each action actually runs.
Workflow & automation · Documented · Documentation checked 2026-10-04
Cloud flows, desktop flows, AI Builder prompts, and Power Platform environments.
Work or school account, licenses, connectors, regional AI capacity, and environment policy affect availability.
The worked example is illustrative and has not been tested in a signed-in product account.
What it does and how it connects
Choose an event, button, or schedule trigger and identify where each action actually runs.
Cloud flows can start from events, a manual button, or a schedule. Desktop flows automate activity on a machine. Connectors provide cloud service operations. An AI step is one component in that process, not a reason to give the entire flow unrestricted authority.
Walkthrough
Use the product named in the access panel. The steps below combine documented controls with the editorial example.
- Sign in with an eligible work / school account and select the intended environment.
- Choose an automated, instant, or scheduled cloud flow for the trigger you need.
- Add a safe sample action with the correct connection and mapped input.
- Test manually and inspect run results before enabling real triggers or writes.
Worked example
An instant sample flow receives a fictional workshop question and returns a private draft. If the required service has a supported connector, use it rather than automating its screen. Desktop RPA introduces separate machine and session requirements.
Illustrative cloud flow
Trigger: manual pilot
Input: request_id and fictional question
Output: private draft / review item
Missing input: stop with explanation
Connected write actions: disabled
Production trigger: added only after review
Inspect and verify the result
Inspect the trigger payload, connection identity, and each action result. Test missing input and a connection failure. Confirm which actions run in the cloud and whether any step depends on a desktop machine.
Limits and troubleshooting
Licensing and connectors differ by environment. Generative actions are documented as preview and should not be assumed universally available. A successful trigger does not establish success of every downstream action.
Data, permissions, and actions
Cloud flows use configured service connections; desktop flows use machine / session authority. Review environment policy, logs, and destination data handling before supplying organizational records.
Related reading
AI Builder: Run a prompt
Map controlled inputs into an AI Builder prompt and inspect its output before taking action.
From an AI prompt to a human-approved action
Place a reviewed draft between AI interpretation and a consequential cloud-flow action.
A desktop draft task with a safe failure path
Create a new draft in a sample desktop application under explicit machine boundaries.
Evaluations: does the system do the job?
Create representative tests for quality, tool behavior, refusal, escalation, and the effects of a change.
Sources and verification
Documentation checked 2026-10-04. Examples are illustrative unless described otherwise.
Types of Power Automate flows · Microsoft