Multi-role

The approval chain, tested end to end

Creator, approver and supplier each run in their own independently authenticated session, so a workflow that spans roles is tested as one journey.

Overview

Multi-role workflows

The workflows most worth testing are the ones that cross roles — a contract created by one person, approved by another, seen by a third — and these are exactly the ones that are painful to script and slow to check by hand. Argus gives each phase its own authenticated session, and names test data per run so parallel agents never collide on a shared environment.

What it does

Journeys, not isolated screens

Each phase runs in a fresh, independently authenticated session.

Creator, approver and supplier journeys tested as one flow.

Run-scoped test data keeps parallel agents from colliding.

Multiple agents run in parallel against one environment.

Login is scripted deterministically, never left to the agent.

How it works

How a multi-role run works

1The scenario declares its phases and their roles
2Each phase opens its own authenticated session
3Test data is named per run to stay isolated
4The whole chain is asserted as one outcome
FAQ

Common questions

How do parallel runs avoid clashing?

Entity names are scoped to the run, so concurrent scenarios never collide on the same test data.

Does each role log in separately?

Yes — each phase gets a fresh, independently authenticated browser session.

Does the agent handle login?

No. Login is performed deterministically by script, so no model calls are spent on a form that never changes.

Explore more

More of what Argus QA does

Get started

See agents test your own application

Request a demo to watch a multi-role regression suite run through your real interface.