Create a new draft in a sample desktop application under explicit machine boundaries.
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.
An illustrative recipe. This website does not run the automation.
The goal and the right kind of system
Create a new draft in a sample desktop application under explicit machine boundaries.
A fixed desktop flow fits a stable UI procedure when no appropriate connector exists. If the process needs broad agent choice, design a separate governed agent rather than treating RPA as one.
Inputs and connected services
Inputs: supported desktop client, approved test machine / session, sample app, invented data, and output location.
Services: Power Automate desktop and the sample application; no production account or outbound connector is required.
The steps
An illustrative sequence to adapt and test before using real data or enabling ongoing work.
- Confirm machine, environment policy, and expected application state.
- Create the sample flow with explicit checks before UI actions.
- Run in the intended session and inspect the saved draft.
- Test missing app / unexpected dialog cases and document manual recovery.
Outputs and human checkpoints
Outputs: new sample draft file and actual run / failure evidence.
Human checkpoints: inspect the actual application and output. Review unattended setup separately before considering production.
Exceptions, permissions, and stopping conditions
Permission boundary: approved machine / sample app only. Wrong window, missing file, or unexpected dialog stops rather than continuing blindly.
Stop before overwrites, production writes, or uncertain UI actions. Keep a person responsible for interruption, cleanup, and recovery.
Adaptable prompt
Illustrative desktop pilot: use only the approved sample app and invented data; verify the expected window before every action; create a new draft; stop on unexpected dialogs or missing input. No overwrite, sending, or production credentials.
Related reading
Desktop automation and machine boundaries
Understand a desktop flow’s machine, session, application, and operational requirements.
Cloud flows, desktop flows, and connectors
Choose an event, button, or schedule trigger and identify where each action actually runs.
Environments, data policies, and governance
Check which connectors may exchange information before activating an AI-enabled flow.
Designing an AI workflow
Make a useful task repeatable with clear inputs, steps, outputs, exceptions, and human checkpoints.
Permissions and meaningful approvals
Decide what a system may read, what it may change, and when someone needs to authorize a concrete action.
Sources and verification
Documentation checked 2026-10-04. Examples are illustrative unless described otherwise.
Create a desktop flow · Microsoft
Desktop flow governance · Microsoft
Types of Power Automate flows · Microsoft
Power Platform data policies · Microsoft