# US 2026/0064569 A1 · Generative pre-training transformed (gpt) based creation of automated script for application testing

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0064569 A1
- Document ID: US20260064569A1
- Status: Published application
- Kind: A1
- Application number: 18/824,239
- Filed: 2024-09-04
- Published: 2026-03-05
- Earliest claimed: 2024-09-04
- Assignees: Sap SE
- Applicants: Sap SE
- Inventors: Ajith G.; Ravikanth MALIPEDDU; Shanavas Madeen S.; Naveen V.
- Web page: https://patentgrep.com/patent/US20260064569A1

## CPC

- G06F 11/3684: {for test design, e.g. generating new test cases}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

In some implementations, there is provided a computer-implemented method that includes determining, via the application function processor, a first suggested step for the test script, the first suggested step for the test script being a step of a workflow for the application obtained from the machine learning model; determining, via the orchestrating model, a first user interface control from a plurality of user interface controls stored within the object repository, and at least a first datum to be entered into a first input field, wherein the first user interface control is associated with a user interface location; and providing, by the orchestrating model and to the user interface, the first suggested step for the test script and a first locator associated with the first user interface control and entering at least the first datum into the first input field. Related systems, methods, and articles of manufacture are also disclosed.

## Claims

### Claim 1

A computer-implemented method comprising:
receiving, at an orchestrating model and from a user via a user interface, a request to create a test script for automated testing of an application of a plurality of applications, the request comprising an application tag identifying the application;
sending, via the orchestrating model, the application tag to an application function processor, an object repository, and a test data container, wherein the application function processor is in communication with a machine learning model trained on at least one database containing one or more functions providing one or more workflows for at least the application;
determining, via the application function processor, a first suggested step for the test script, the first suggested step for the test script being a step of a workflow for the application obtained from the machine learning model;
determining, via the orchestrating model, a first user interface control from a plurality of user interface controls stored within the object repository, and at least a first datum to be entered into a first input field, wherein the first user interface control is associated with a user interface location; and
providing, by the orchestrating model and to the user interface, the first suggested step for the test script and a first locator associated with the first user interface control and entering at least the first datum into the first input field.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, further comprising, upon receiving an indication that at least the first datum has been entered into the first input field,
determining, by the application function processor, a second suggested step for the test script, a second user interface control and at least a second datum to be entered into a second input field;
providing, by the orchestrating model and to the user interface, the second suggested step and a second locator corresponding to the second user interface control; and
entering at least the second datum into the second input field.

### Claim 3 (depends on 2)

The computer-implemented method of claim 2, wherein the first user interface control and the second user interface control are provided by the object repository.

### Claim 4 (depends on 2)

The computer-implemented method of claim 2, wherein the application function processor determines the first suggested step for the test script and the second suggested step for the test script using the machine learning model comprising a large language model.

### Claim 5 (depends on 4)

The computer-implemented method of claim 4 wherein at least a first document of a plurality of documents lists the one or more functions of the application.

### Claim 6 (depends on 5)

The computer-implemented method of claim 5, wherein the first suggested step for the test script and the second suggested step for the test script are steps comprising a guided flow described by a working model of at least the first document of the plurality of documents.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein the first datum to be entered into the first input field is provided by a test data container.

### Claim 8 (depends on 1)

The computer-implemented method of claim 1, wherein the test script is provided to test the application, and record testing, using the test script, of the application.

### Claim 9

A system comprising:
at least one processor; and
at least one memory including instructions which when executed by the at least one processor causes operations comprising:
receiving, at an orchestrating model and from a user via a user interface, a request to create a test script for automated testing of an application of a plurality of applications, the request comprising an application tag identifying the application;
sending, via the orchestrating model, the application tag to an application function processor, an object repository, and a test data container, wherein the application function processor is in communication with a machine learning model trained on at least one database containing one or more functions providing one or more workflows for at least the application;
determining, via the application function processor, a first suggested step for the test script, the first suggested step for the test script being a step of a workflow for the application obtained from the machine learning model;
determining, via the orchestrating model, a first user interface control from a plurality of user interface controls stored within the object repository, and at least a first datum to be entered into a first input field, wherein the first user interface control is associated with a user interface location; and
providing, by the orchestrating model and to the user interface, the first suggested step for the test script and a first locator associated with the first user interface control and entering at least the first datum into the first input field.

### Claim 10 (depends on 9)

The system of claim 9, further comprising, upon receiving an indication that at least the first datum has been entered into the first input field,
determining, by the application function processor, a second suggested step for the test script, a second user interface control and at least a second datum to be entered into a second input field;
providing, by the orchestrating model and to the user interface, the second suggested step and a second locator corresponding to the second user interface control; and
entering at least the second datum into the second input field.

### Claim 11 (depends on 10)

The system of claim 10, wherein the first user interface control and the second user interface control are provided by the object repository.

### Claim 12 (depends on 10)

The system of claim 10, wherein the application function processor determines the first suggested step for the test script and the second suggested step for the test script using the machine learning model comprising a large language model.

### Claim 13 (depends on 12)

The system of claim 12, wherein at least a first document of a plurality of documents lists the one or more functions of the application.

### Claim 14 (depends on 13)

The system of claim 13, wherein the first suggested step for the test script and the second suggested step for the test script are steps comprising a guided flow described by a working model of at least the first document of the plurality of documents.

### Claim 15 (depends on 9)

The system of claim 9, wherein the first datum to be entered into the first input field is provided by a test data container.

### Claim 16 (depends on 9)

The system of claim 9, wherein the test script is provided to test the application, and record testing, using the test script, of the application.

### Claim 17

A non-transitory computer-storage medium including instructions which when executed by at least one processor causes operations comprising:
receiving, at an orchestrating model and from a user via a user interface, a request to create a test script for automated testing of an application of a plurality of applications, the request comprising an application tag identifying the application;
sending, via the orchestrating model, the application tag to an application function processor, an object repository, and a test data container, wherein the application function processor is in communication with a machine learning model trained on at least one database containing one or more functions providing one or more workflows for at least the application;
determining, via the application function processor, a first suggested step for the test script, the first suggested step for the test script being a step of a workflow for the application obtained from the machine learning model;
determining, via the orchestrating model, a first user interface control from a plurality of user interface controls stored within the object repository, and at least a first datum to be entered into a first input field, wherein the first user interface control is associated with a user interface location; and
providing, by the orchestrating model and to the user interface, the first suggested step for the test script and a first locator associated with the first user interface control and entering at least the first datum into the first input field.

### Claim 18 (depends on 17)

The non-transitory computer-storage medium of claim 17, further comprising, upon receiving an indication that at least the first datum has been entered into the first input field,
determining, by the application function processor, a second suggested step for the test script, a second user interface control and at least a second datum to be entered into a second input field;
providing, by the orchestrating model and to the user interface, the second suggested step and a second locator corresponding to the second user interface control; and
entering at least the second datum into the second input field.

### Claim 19 (depends on 18)

The non-transitory computer-storage medium of claim 18, wherein the first user interface control and the second user interface control are provided by the object repository.

### Claim 20 (depends on 18)

The non-transitory computer-storage medium of claim 18, wherein the application function processor determines the first suggested step for the test script and the second suggested step for the test script using the machine learning model comprising a large language model.
