Information processing system, information processing method, and program

Abstract

[Problem] To support automation of a test. [Solution] This information processing system is characterized by including: a scenario generation unit for generating a scenario of a test by giving a first prompt instructing generation of the scenario to a large-scale language model on the basis of a test case describing an outline of the test; and a script generation unit for generating a script by giving a second prompt to the large-scale language model, the second prompt instructing generation of the script that causes a computer to execute the test on the basis of the scenario.

Description

Information processing system, information processing method and program

[0001] The present invention relates to an information processing system, an information processing method, and a program.

[0002] Patent Document 1 discloses generating test code according to test code generation rules.

[0003] Japanese Patent Application Laid-Open No. 2007-066204

[0004] However, Patent Document 1 requires design specifications to which the rules can be applied.

[0005] The present invention has been made in view of the above background, and aims to provide a technique that can support test automation.

[0006] The main invention of the present invention for solving the above problem is an information processing system comprising: a scenario generation unit that generates a scenario for a test based on a test case that describes an overview of the test by providing a first prompt to a large-scale language model, the first prompt instructing the model to generate the scenario; and a script generation unit that generates the script by providing a second prompt to the large-scale language model instructing the model to generate a script that causes a computer to execute the test based on the scenario.

[0007] Other problems and solutions disclosed in this application will be made clear in the section on preferred embodiments of the invention and the drawings.

[0008] According to the present invention, it is possible to support the automation of testing.

[0009] It is a diagram showing an example of the overall configuration of an information processing system. It is a diagram showing an example of the hardware configuration of a management server 2. It is a diagram showing an example of the software configuration of a management server 2. It is a diagram showing an example of an inquiry screen. It is a diagram explaining the operation of the management server 2. It is a diagram explaining the script generation process.

[0010] <System Overview> An information processing system according to one embodiment of the present invention will be described below. The information processing system according to this embodiment supports program testing. In this embodiment, it is assumed that the information processing system supports testing of web applications. The information processing system according to this embodiment uses a machine learning model capable of generating responses to instructions to infer elements to be tested (hereinafter referred to as target elements) in screen data (HTML) output by the web application and operations on the target elements, and proposes the elements and their operations to be tested to the user. In this embodiment, the machine learning model is assumed to be a large language model (LLM). Known examples of LLMs include, but are not limited to, GPT. Furthermore, the machine learning model may be a learning model other than an LLM, as long as it is capable of executing the task of generating responses to instructions (prompts).

[0011] 1 is a diagram showing an example of the overall configuration of an information processing system. The information processing system of this embodiment is configured to include a management server 2. The management server 2 is communicably connected to a user terminal 1, a target server 3, and a generation server 4 via a communication network. The communication network is, for example, the Internet, and is constructed using a public telephone network, a mobile phone network, a wireless communication path, Ethernet (registered trademark), or the like.

[0012] The user terminal 1 is a computer operated by a user, and may be, for example, a smartphone, a tablet computer, or a personal computer.

[0013] The management server 2 is a computer that supports testing. The management server 2 may be a general-purpose computer such as a workstation or a personal computer, or may be logically realized by cloud computing.

[0014] The target server 3 is a computer that executes a program to be tested. In this embodiment, the target server 3 is a server that executes a so-called web application. The target server 3 may be a general-purpose computer such as a workstation or a personal computer, or may be logically realized by cloud computing.

[0015] The generation server 4 is a computer that performs processing using the LLM. The generation server 4 may be a general-purpose computer such as a workstation or a personal computer, or may be logically realized by cloud computing. The generation server 4 includes the LLM and can, for example, provide an API, receive prompts from outside, provide the received prompts to the LLM to generate answers, and respond with the answers.

[0016] <Management Server> FIG. 2 is a diagram illustrating an example of the hardware configuration of the management server 2. Note that the illustrated configuration is an example, and other configurations may also be used. The management server 2 includes a CPU 201, memory 202, storage device 203, communication interface 204, input device 205, and output device 206. The storage device 203 stores various data and programs, and is, for example, a hard disk drive, solid state drive, or flash memory. The communication interface 204 is an interface for connecting to a communication network, and is, for example, an adapter for connecting to Ethernet (registered trademark), a modem for connecting to a public telephone network, a wireless communication device for wireless communication, or a USB (Universal Serial Bus) connector or RS232C connector for serial communication. The input device 205 is, for example, a keyboard, mouse, touch panel, button, microphone, or the like for inputting data. The output device 206 is, for example, a display, printer, speaker, or the like for outputting data. Each functional unit of the management server 2 described below is realized by the CPU 201 reading a program stored in the storage device 203 into the memory 202 and executing it, and each storage unit of the management server 2 is realized as part of the storage area provided by the memory 202 and the storage device 203.

[0017] 3 is a diagram illustrating an example of the software configuration of the management server 2. The management server 2 includes a document reception unit 211, a test case generation unit 212, a scenario generation unit 213, a script generation unit 214, an inquiry unit 215, and a script execution unit 216.

[0018] <Function Unit> The document receiving unit 211 receives documents related to the test target (program). The documents may be specifications (product specifications) related to the test target. In this embodiment, it is assumed that the specifications do not include detailed input / output data specifications or screen usage, but these detailed specifications may be included. The documents may include information other than the specifications (additional information). The documents may include, for example, product specifications for functions other than the test target. The documents may also include, for example, documents related to programs other than product specifications. The documents may include, for example, operating procedures and help documents. The documents may include, for example, image information other than that included in the specifications, such as wireframes, conceptual diagrams, and design documents. The documents may include, for example, documents related to laws, regulations, rules, etc. with which the test target program or the generated test cases and test scenarios must comply. The documents may also include other information written in natural language that the test target program or the test cases and test scenarios must comply. The documents may include text data, image data, etc. related to the test target.

[0019] The test case generation unit 212 generates test cases based on the documents received by the document reception unit 211. A test case defines what is to be tested. A test case can be, for example, a description of a test outline. A test case can be, for example, the outline (epic) of a use case. A test case can also be, for example, the outline (epic) of a user story. The test case generation unit 212 can generate test cases by providing a prompt (test case generation prompt, third prompt) to the LLM instructing it to generate test cases based on the documents.

[0020] The test case generation prompt can include instructions to generate a test case, documentation (specifications and additional information), and additional information written in natural language (e.g., test case granularity). The prompt can also include conditions required for the test case. A list of test cases indicating the content (overview) to be tested can be obtained from the LLM. In this embodiment, the test case generation unit 212 transmits this test case generation prompt to the generation server 4, and the generation server 4 can receive answers (test cases) generated in response to the prompt. The test case generation prompt can also include instructions to specify the characteristics expected of the test case. For example, instructions such as "cover all major functions" and "include positive and negative cases" can be included.

[0021] An example of a test case generation prompt is shown below:

[0022] Create test cases for the web application based on the following information. Specifications: (Contents of the web application specifications) Expected characteristics: ・Encompass major functions ・Include positive and negative cases ・Covers not only functional requirements but also non-functional requirements (performance, security, etc.) ・Each test case clearly states the test target, procedures, and expected results Test case format: ・Output in the following format ・Title ・Description ・Preconditions ・Procedures ・Expected results

[0023] Note that the information to be included in the test case generation prompt is not limited to the above example. The information to be included in the prompt can be changed as appropriate depending on the characteristics of the application to be tested and the purpose of the test. For example, when generating test cases specialized for security testing, information about the type of vulnerability and attack method can be included in the prompt.

[0024] The scenario generation unit 213 generates a test scenario based on a test case (which describes an overview of the test). The scenario is assumed to be written in Gherkin notation. The scenario generation unit 213 can generate a scenario by providing a prompt (scenario generation prompt, first prompt) to the LLM, which instructs the LLM to generate a test scenario based on the test case.

[0025] The scenario generation prompt can include test cases and documents (specifications and additional information). The scenario generation prompt can also include conditions required for the test scenario. A test scenario written in Gherkin notation can be obtained from the LLM. In this embodiment, the scenario generation unit 213 can send this scenario generation prompt to the generation server 4 and receive an answer (scenario) generated in response to the prompt from the generation server 4. The scenario generation prompt can also specify the characteristics expected of the test scenario. For example, instructions such as "cover the user's main operational flow" and "include exceptional cases" can be included.

[0026] A specific example of a scenario generation prompt is shown below.

[0027] Create a test scenario for the web application based on the following information. Test cases: ・Confirm that the user can log in ・Confirm that the user can search for products ・Confirm that the user can purchase products Specifications: (Contents of the web application specifications) Expected characteristics: ・Encompass the main operation flow ・Include exceptional cases (login failure, product not found, etc.) ・Specify the expected results for each operation Scenario format: ・Write in Gherkin notation

[0028] The script generation unit 214 generates a script that causes a computer to execute a test based on the scenario. The script can be, for example, a script that can be executed by Playwright. The script generation unit 214 can generate the script by providing a prompt (script generation prompt, second prompt) to the LLM instructing the LLM to generate a script that causes a computer to execute a test based on the scenario.

[0029] The script generation prompt may include screen data of the program to be tested. In this embodiment, since the program to be tested is assumed to be a web application provided by the target server 3, the screen data may be, for example, text data written in HTML. Furthermore, instead of or in addition to HTML data, the screen data may include, for example, screenshot information of a screen displayed in a web browser. Furthermore, the script generation prompt may include an instruction to inquire about the identification of an element to be operated upon when generating the script, if such identification is necessary. Furthermore, the script generation prompt may include an instruction to output user input, if such input is required when generating the script. Furthermore, the script generation prompt may include instructions to specify characteristics expected of the test script. For example, instructions such as "use Playwright" or "use a page object pattern" may be included. In this embodiment, the script generation unit 214 transmits this script generation prompt to the generation server 4, and the generation server 4 can receive an answer (script) generated in response to the prompt. If the response from the LLM includes a script (program code), the script generation unit 214 can add the script to the final script.

[0030] An example of a script generation prompt is shown below.

[0031] Create a test script for the web application in TypeScript based on the following information. Test scenario: (Test scenario written in Gherkin notation) Screen information: (HTML of the screen to be tested) Expected characteristics: - Use Playwright - Use the page object pattern - Use the expect function of Playwright for assertions

[0032] Note that the test scripts generated in this embodiment are not limited to scripts using Playwright. The script generation unit 214 can generate various types of test scripts depending on the type of application to be tested, the purpose of the test, and available test automation frameworks.

[0033] For example, when performing an E2E test on a web application, the script generation unit 214 can generate a script using a framework for automatically operating a web browser, such as Playwright, Selenium, Cypress, Puppeteer, etc. Note that the script for operating a web browser is not limited to these.

[0034] Furthermore, when testing a mobile application, the script generation unit 214 can generate a script using a framework for automatically operating an application on a mobile device, such as Apppium, Espresso, or XCUITest. Note that scripts for operating mobile applications are not limited to these.

[0035] Furthermore, when testing a desktop application, the script generation unit 214 can generate a script using a framework for automatically operating a desktop application, such as WinAppDriver, ApppiumDesktop, AutoIT, etc. Note that scripts for operating desktop applications are not limited to these.

[0036] In addition, the script generation unit 214 of this embodiment is not limited to using a single automation framework when generating a test script. For example, when performing a test that combines a web application and an API, the script generation unit 214 can generate a script that combines multiple frameworks (for example, Playwright and Postman) to automate both the web browser and the API.

[0037] The query unit 215 queries the user for information necessary for generating a script. When the query unit 215 cannot determine an element to be operated in the script, it can query the user for the element. When the query unit 215 cannot generate input data for an element in the script, it can query the user for input data. The query unit 215 can query the user in response to a response from the LLM. When the response from the LLM in response to the script generation prompt includes content inquiring about an element, the query unit 215 can query the element to be operated. When the response from the LLM in response to the script generation prompt includes content requesting input data, the query unit 215 can cause the user to input input data (text data). For example, the query unit 215 can transmit screen data including an input form to the user terminal 1 and cause the user terminal 1 to specify an element and input input data from the input form.

[0038] The query unit 215 can present a query screen to the user when querying the user about elements or input data. The query screen displays the screen of the application to be tested, and allows the user to select a specific element or enter input data.

[0039] 4 is a diagram showing an example of an inquiry screen. The inquiry screen is displayed on the user terminal 1. A screenshot 601 of the application screen to be tested is displayed on the inquiry screen 600. On the screenshot 601, the user can select an element that the user wants to operate with the script using a mouse, a touch panel, or the like.

[0040] The query screen 600 further displays an element information field 602 for displaying information about the selected element, an operation type selection field 603 for selecting the type of operation for the element, and a data input field 604 for inputting data to be input into the element. The user can confirm the selected element in the element information field 602, select an operation for the element (click, text input, etc.) in the operation type selection field 603, and enter input data in the data input field 604 as necessary.

[0041] The inquiry screen 600 further displays a Confirm button 605 for confirming the results of the selection or input, and a Reset button 606 for redoing the selection or input. By pressing the Confirm button 605, the user can transmit the results of the selection or input to the inquiry unit 215. By pressing the Reset button 606, the contents of the selection or input can be cleared.

[0042] The inquiry unit 215 can transfer element information and input data acquired from the user via the inquiry screen 600 to the script generation unit 214. The script generation unit 214 can reflect the acquired information in the script.

[0043] The script execution unit 216 executes the script generated by the script generation unit 214. The script execution unit 216 can output the execution result of the script as a test result.

[0044] <Operation> FIG. 5 is a diagram illustrating the operation of the management server 2.

[0045] The management server 2 receives various documents (materials) related to the test subject (S301), and generates a test case by providing a test case generation prompt including the received materials to the LLM (S302). The management server 2 receives the selection of one test case (S303), and generates a scenario by providing a scenario generation prompt including the selected test case to the LLM (S304). The management server 2 generates a script by providing a script generation prompt including the scenario to the LLM (S305).

[0046] As described above, according to the information processing system of this embodiment, test cases, scenarios, and scripts can be generated simply by providing materials related to the test subject.

[0047] FIG. 6 is a diagram illustrating the script generation process.

[0048] The management server 2 reads the next block (one processing unit) from the scenario in order from the beginning (S321), and gives a script generation prompt including the read block to the LLM to obtain a response (S322).

[0049] If the obtained answer is not a script (S323: NO), or if the answer is an inquiry about an element (S324: YES), the management server 2 inquires the user about the element to be operated (S325), includes the element specified by the user in the script generation prompt (S326), gives the script generation prompt to the LLM again to obtain an answer (S327), and proceeds to step S323.

[0050] If the acquired answer is not a script (S323: NO), or if the answer is a query for input data (S324: NO), the management server 2 queries the user for the input data (S328), includes the input data entered by the user in the script generation prompt (S329), proceeds to step S327, again sends the script generation prompt to the LLM to acquire an answer, and proceeds to step S323.

[0051] If the response from the LLM is a script (S323: YES), the management server 2 adds the returned script to the final test script (S330). The management server 2 repeats the process from step S321 until all scenarios have been read (S331: NO).

[0052] As described above, according to the information processing system of this embodiment, it is possible to generate a script that automates operations on a specific screen from an abstract scenario using LLM.

[0053] Although the present embodiment has been described above, the above embodiment is intended to facilitate understanding of the present invention and is not intended to limit the present invention. The present invention may be modified or improved without departing from the spirit thereof, and equivalents thereof are also included in the present invention.

[0054] For example, the processing by each of the functional units of the management server 2 described above may be performed by any of the functional units. Also, a different functional unit that performs part of the processing by each of the functional units described above may be added. Also, the functional units of the management server 2 may be distributed across multiple computers.

[0055] Furthermore, the information stored in each storage unit of the management server 2 may be stored in any of the storage units. That is, the information stored in the above-mentioned multiple storage units may be stored in one storage unit, or part of the information stored in one of the above-mentioned storage units may be stored in another storage unit.

[0056] In this embodiment, the applications to be tested are not limited to web applications. The information processing system of this embodiment can support automated testing of various types of applications.

[0057] For example, the information processing system of this embodiment can be applied to the automation of testing of mobile applications. In this case, the document receiving unit 211 can receive specifications and design documents for the mobile application. Furthermore, the script generating unit 214 can analyze the screen of the mobile application and identify elements on the screen.

[0058] The information processing system of this embodiment can also be applied to the automation of testing of desktop applications. In this case, the document receiving unit 211 can receive specifications and design documents for the desktop application. The script generating unit 214 can analyze the GUI of the desktop application and identify elements such as windows and buttons.

[0059] Furthermore, the information processing system of this embodiment can be applied to the automation of tests for various types of software systems, such as embedded systems and IoT devices. In this case, the document receiving unit 211 can receive system specifications, design documents, communication protocol definition documents, etc. Furthermore, the script generating unit 214 can generate test scripts based on communication procedures and data formats with the system.

[0060] <Variation 1> In addition, in the above embodiment, the purpose of the test was not particularly limited when generating test cases and scenarios, but appropriate test cases and scenarios may be generated depending on the purpose of the test.

[0061] For example, the test case generator 212 can include information indicating the purpose of the test in the test case generation prompt. The purpose of the test can be, for example, a regression test, an acceptance test, an integration test, a system test, a performance test, a security test, etc. The test case generator 212 can instruct the LLM to generate test cases that satisfy the specified purpose of the test.

[0062] Similarly, the scenario generator 213 can include information indicating the purpose of the test in the scenario generation prompt, and can instruct the LLM to generate a test scenario that satisfies the specified test purpose.

[0063] For example, if the purpose is regression testing, the test case generation unit 212 and the scenario generation unit 213 can instruct the LLM to generate test cases and scenarios that comprehensively test the changed functions while satisfying the specifications of the program before the change.

[0064] Furthermore, when the purpose is acceptance testing, the test case generation unit 212 and the scenario generation unit 213 can instruct the LLM to generate test cases and scenarios that cover the user's main use cases.

[0065] <Modification 2> In the above embodiment, the application to be tested is assumed to be a web application, but this is not limiting. The application to be tested may be another type of application, such as a mobile application or a desktop application.

[0066] When a mobile application is to be tested, screen data can be, for example, a screenshot image of the application or XML data constituting the screen. The script generation unit 214 can generate a test script for the mobile application based on this screen data. For example, the test script can be generated as an executable script using a test framework for mobile applications such as Apppium.

[0067] When a desktop application is to be tested, screen data can be, for example, a screenshot image of the application or property information of screen elements obtained from the UI automation API of the OS. The script generation unit 214 can generate a test script for the desktop application based on this screen data. The test script can be generated using a test framework for desktop applications, such as WinAppDriver or PyAutoGUI.

[0068] <Modification 3> In the above embodiment, the GUI (Graphical User Interface) of the application is assumed to be the test target, but this is not limiting. The test target may also be an API (Application Programming Interface) provided by the application.

[0069] When an API is to be tested, the document receiving unit 211 can receive the API specification. The API specification describes, for example, the names, arguments, return values, error codes, etc. of the methods provided by the API. The test case generation unit 212 can generate test cases for testing the API based on the API specification. For example, the test cases can be generated to cover normal and abnormal patterns for each method.

[0070] The scenario generation unit 213 can generate a test scenario for testing the API based on the generated test cases. The test scenario can include, for example, the order in which the test cases are executed, test data settings, expected results, and the like.

[0071] The script generation unit 214 can generate a test script that sends an HTTP request to an API and verifies the response based on the generated test scenario. The test script can be a script that can be executed by an API test tool such as JMeter or Postman.

[0072] <Modification 4> In the above embodiment, the function of generating test cases, test scenarios, and test scripts using the LLM has been described, but this is not limiting. The LLM may also be used to analyze the results of test execution.

[0073] For example, the management server 2 may be newly equipped with a test result analysis unit. The test result analysis unit analyzes the test results obtained as a result of the script execution unit 216 executing the test script. The test results include, for example, an execution log for each test case, detected errors, performance measurements, and the like.

[0074] The test result analysis unit analyzes the test results, and if a problem is detected, it can provide the LLM with an analysis prompt to identify the cause of the problem. The analysis prompt can include, for example, information about the test results and information about the application being tested. The LLM infers the cause of the problem based on the provided information and returns the result to the test result analysis unit.

[0075] The test result analysis unit can also provide the LLM with an analysis prompt that instructs it to propose a fix for the identified problem cause. The analysis prompt can include, for example, information about the problem cause and information about the application being tested. The LLM generates a fix based on the provided information and returns the result to the test result analysis unit. The proposed fix can include, for example, changes to the source code or changes to the settings.

[0076] The test result analysis unit can display the causes of problems identified by the LLM and the proposed fixes, for example, on the screen of the user terminal 1. This allows the user who performed the test to refer to the causes of problems and the proposed fixes and efficiently fix the application's defects.

[0077] <Modification 5> In the above embodiment, the function of generating information about a test using an LLM has been described, but this is not limiting. An LLM may be used to generate a document about an application to be tested.

[0078] For example, the management server 2 may be newly provided with a document generation unit. The document generation unit generates documents such as user manuals and tutorials related to the application to be tested.

[0079] The document generation unit can acquire information such as the specifications of the application to be tested, GUI screen data, and API specifications. Based on the acquired information, the document generation unit can provide the LLM with a document generation prompt that instructs it to generate a user manual or tutorial. The document generation prompt can instruct the LLM to include, for example, an overview of the application's functions, operating procedures for major screens, explanations of input forms, and examples of API usage.

[0080] Based on the information and prompts provided, the LLM generates a draft of the user manual or tutorial and returns the result to the document generator, which can then make adjustments to the draft, such as formatting or adding figures and tables.

[0081] The document generator can also use the LLM to proofread a user manual or tutorial manuscript. For example, the LLM can be given a proofreading prompt to check whether the manuscript is consistent with the specifications. The LLM can compare the manuscript with the specifications and point out any inconsistencies.

[0082] The document generation unit can display the generated user manual or tutorial on the screen of the user terminal 1, for example, or output it to a file, thereby enabling application developers to create documents efficiently.

[0083] <Variation 6> In the above embodiment, it is assumed that a script is generated based on a scenario prior to test execution, but it is also possible to generate a script while a test is being executed. In this case, the script generation unit 214 may include, in the script generation prompt, in addition to the scenario description, the screen of the application to be tested (which can be described in HTML in the case of a Web application) and status information of the test target page corresponding to the scenario description, in order from the beginning of the generated test scenario. When a script is generated by LLM, the script execution unit 216 may execute the script step by step and update the status information.

[0084] In this case, the script generation unit 214 may inquire of the user each time about the element to be operated and the operation content (and input data in the case of an input operation). The script generation unit 214 can cause the LLM to generate a script by including the element specified by the user and the operation for that element (and input data in the case of an input operation) in a script generation prompt.

[0085] <Disclosures> The present disclosure also includes the following configurations. [Item 1] An information processing system comprising: a scenario generation unit that generates the scenario by providing a large-scale language model with a first prompt instructing the model to generate a scenario for the test based on a test case describing an overview of the test; and a script generation unit that generates the script by providing a second prompt instructing the large-scale language model to generate a script for causing a computer to execute the test based on the scenario. [Item 2] The information processing system according to item 1, comprising: a document acceptance unit that accepts a document related to a target of the test; and a test case generation unit that generates the test case by providing a third prompt instructing the large-scale language model to generate the test case based on the document. [Item 3] The information processing system according to item 1, further comprising: a query unit that, when an element to be operated in the script cannot be determined, queries a user about the element. [Item 4] The information processing system according to item 1, further comprising a query unit that queries a user for input data when input data for an element cannot be generated in the script. [Item 5] The information processing system according to item 1, further comprising a script execution unit that executes the script. [Item 6] An information processing method, characterized in that a computer executes the steps of: generating a scenario for the test by providing a large-scale language model with a first prompt that instructs the large-scale language model to generate the scenario based on a test case that describes an overview of the test; and generating the script by providing the large-scale language model with a second prompt that instructs the computer to generate a script that causes the computer to execute the test based on the scenario.[Item 7] A program for causing a computer to execute the steps of: generating a scenario for a test by providing a large-scale language model with a first prompt that instructs the model to generate a scenario for the test based on a test case that describes an overview of the test; and generating the script by providing a second prompt that instructs the model to generate a script that causes the computer to execute the test based on the scenario.

[0086] 1 User terminal 2 Management server 3 Target server 4 Generation server

Claims

1. An information processing system comprising: a scenario generation unit that generates a scenario by providing a large-scale language model with a first prompt instructing it to generate a scenario for a test based on a test case that describes an overview of the test; and a script generation unit that generates the script by providing the large-scale language model with a second prompt instructing it to generate a script that causes a computer to execute the test based on the scenario.

2. An information processing system according to claim 1, comprising: a document receiving unit that receives documents related to the subject of the test; and a test case generation unit that generates the test cases by providing a third prompt to the large-scale language model instructing it to generate the test cases based on the documents.

3. An information processing system according to claim 1, further comprising an inquiry unit that, when the element to be operated cannot be determined in the script, inquires of the user about the element.

4. An information processing system according to claim 1, further comprising an inquiry unit that, when input data for an element cannot be generated in the script, inquires of a user about the input data.

5. An information processing system according to claim 1, further comprising a script execution unit that executes the script.

6. An information processing method characterized by being executed by a computer, the method comprising: generating a scenario for a test by providing a first prompt to a large-scale language model instructing the model to generate a scenario for the test based on a test case that describes an overview of the test; and generating a script by providing a second prompt to the large-scale language model instructing the model to generate a script that causes the computer to execute the test based on the scenario.

7. A program for causing a computer to execute the steps of: generating a scenario for a test by providing a large-scale language model with a first prompt instructing the model to generate a scenario for the test based on a test case that describes an overview of the test; and generating a script by providing a second prompt instructing the large-scale language model to generate a script that causes the computer to execute the test based on the scenario.

Images