Manual test cases in. Certify automation out.
AutoTestr drives your SAP and web applications through the spreadsheets you already have, and hands back the capture your own recorder would have made.
- WorkbookThe Excel test cases your testers already maintain
- Live applicationDriven step by step from a runner inside your network
- Certify captureCapture XML and the process-export ZIP, ready to import
What is MindMap AutoTestr?
MindMap AutoTestr turns the manual test cases you already keep in Excel into Worksoft Certify automation. A runner inside your network drives the live application through each step and records how Certify should identify every object, by executing your own Certify (XF) recognition rules. The output is Certify Business Capture XML and the Certify process-export ZIP.
- Object identity comes from executing your XF rule tables, not from guessing.
- A step it cannot resolve is flagged for review, never replaced by a plausible substitute.
- Credentials never leave the runner.
- No language model is used unless a run enables one; the runner kit carries none.
The test cases are written. Recording them is the bottleneck.
The cases are already written
Certify teams own hundreds of manual test cases in spreadsheets: the sequence, the field each step touches, the data. Nobody has the time to record them.
The recorder moves at the speed of people
The recorder is right for one flow. Someone has to perform each case by hand while it records, so the backlog shrinks one tester at a time.
Guessed identity doesn't run
Generators that invent how to find an object produce objects Certify cannot find. Identity has to come from your own Certify configuration.
Three steps from spreadsheet to Certify.
Bring the workbook
Your existing spreadsheets, as they are. Column layouts differ from team to team; headers are matched by meaning, and field lists are expanded.
It drives the live application
A runner inside your network walks each case forward through SAP Fiori, WebGUI, WebDynpro, SAP GUI or web screens, reaching screens that only exist after earlier steps.
Review, then import into Certify
Each step shows the element chosen, its confidence and the runners-up. Confirm or override, then take the Capture XML and the Certify process-export ZIP.
Measured against the capture a person made by hand.
- objects identical to the customer's own hand capture
- 17 of 17
- Advance Order
- steps identical to the customer's own hand capture
- 19 of 19
- Advance Order
- identity strings byte-identical to Worksoft's own recorder
- 61 of 62
- reference captures
On Advance Order, the workbook was driven end to end and the result compared with the capture the customer's own team had recorded by hand from the same workbook. The identity strings were compared with reference captures made by Worksoft's recorder. We don't publish time-saved or accuracy figures yet: those get measured on your cases, in the pilot.
The runner always stays in your network.
The application under test lives inside your network, so the runner that drives it does too. The editions differ in where the control plane runs and what may leave.
Cloud
Portal, review and generation hosted by MindMap. Your Windows runner connects out over HTTPS only. Credentials stay on the runner. Multi-tenant: each customer has its own sign-in, sees only its own runs, and has them driven only by its own runners. Single sign-on through your identity provider, with roles for administrators, reviewers and authors.
Sovereign
The whole control plane inside your network, single-tenant, with no outbound connection. Updates and licence arrive as signed files you apply.
Standalone kit
One desktop, no portal: the runner converts and writes the XML and ZIP to a folder. The smallest footprint, and a practical vehicle for a pilot.
The Cloud edition is live: request access, or sign in if your workspace is already set up.
Rules first. The model is off unless you turn it on.
AutoTestr resolves steps with rules that execute your own Certify configuration. A language model is only a fallback for steps the rules cannot settle, it is off unless a run enables it, and the runner kit that works inside your network carries no model client at all. If you opt in, MindMap Sovereign AI, an open-weight model on MindMap's own GPU host or inside your perimeter, gives a second reading of each step and explains failed runs. Its answers are advisory: anything a model chose is marked and goes to review.
What Certify leads and security teams ask first
What does MindMap AutoTestr produce?
Can it be accurate on our test cases?
Which applications does it drive?
What AI model sees our data?
What happens to our passwords?
Start with a census of your workbooks.
One week, parse-only: every workbook you offer, classified by application, with what each sheet is missing. Useful on its own, and it sizes the pilot.
Worksoft and Certify are trademarks of their owner. MindMap AutoTestr is a MindMap Digital product; naming Certify describes the files AutoTestr produces and does not imply endorsement by Worksoft. Capability status words on these pages (Built, Built not run live, Roadmap) describe the current release.