Device and method for creating technical document for software, and program therefor

Abstract

To realize the efficiency by application of a generative AI model in a method for creating a technical document necessary for development of software.SOLUTION: First, a device 100 receives explanation data representing the explanation of software to be developed (S201). The device 100 requests a first generative AI model to generate a plurality of use cases on the basis of the explanation data (S202), and the first generative AI model generates the use cases and transmits them to the device 100 (S203 and S204). Next, the device 100 requests a second generative AI model to generate one or more pieces of business logic usable by a use case on the basis of the use case (S205), and the second generative AI model generates the pieces of business logic and transmits them to the device 100 (S206 and S207). Then, for each of them, the device 100 requests a third generative AI model to generate a corresponding API document (S208).SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to an apparatus, a method, and a program therefor for creating software technical documents.

Background Art

[0002] There are various methods for software development, and among them, use cases are often created in the software requirements analysis stage. Here, the "use case" refers to the behavior of the system or its description in response to input from the user interface (UI), and is often described by business-side members such as business analysts, product owners, and product managers who can understand and express the business requirements for the software to be developed.

[0003] The created use cases are used by development-side members such as system architects, technical analysts, and technical leads to create technical documents with more technical details, and the necessary coding is performed based on the created technical documents.

Summary of the Invention

Problems to be Solved by the Invention

[0004] Thus, in software development, starting from customer hearings, involving a large number of members and a large number of processes, efficiency has always been required. The inventors have found that by using a generative AI model whose capabilities have been significantly improved in recent years, a significant improvement in efficiency can be achieved particularly in the process of creating technical documents.

[0005] The present invention has been made in view of such points, and the problem is to achieve efficiency improvement by applying a generative AI model in an apparatus, a method, or a program therefor for creating technical documents required for software development.

[0006] As used in this specification, the term "AI model" refers to a machine learning model that has been trained to be able to predict an output for an input, and the term "generative AI model" refers to a large language model (LLM) that has been trained using text data to be able to generate an output for an input.

Means for Solving the Problem

[0007] To achieve such an object, a first aspect of the present invention is a method for creating a technical document of software, including: a step of obtaining description data in which the description of the software is expressed, where the description data includes at least one of one or more sentences and one or more images; a first request step of requesting a first generative AI model to generate one or more use cases that occur in the software based on the description data; a second request step of, for at least any one of the one or more use cases, requesting a second generative AI model to generate one or more business logics corresponding to the use case; and a third request step of, for at least any one of the one or more business logics, requesting a third generative AI model to generate API documentation corresponding to the business logic.

[0008] Further, a second aspect of the present invention is the method of the first aspect, where the description data includes one or more images, and the one or more images include a plurality of images in which a series of screens are represented.

[0009] Further, a third aspect of the present invention is the method of the first aspect, where the description data includes one or more images, and the one or more images include a sequence diagram showing the behavior of the software.

[0010] Further, a fourth aspect of the present invention is the method of any one of the first to third aspects, where the second request includes a template of a business logic having items of input, logic, and response.

[0011] Further, a fifth aspect of the present invention is the method according to any one of the first to fourth aspects, wherein the third requirement includes a template of an API document having items of a request type, parameters, an endpoint, an API response code, and a response body.

[0012] Also, a sixth aspect of the present invention is the method according to the fifth aspect, wherein the template of the API document further has items of corresponding business logic.

[0013] Also, a seventh aspect of the present invention is the method according to the fourth aspect, wherein the second requirement includes an instruction to generate the items of the input using column names included in the database design of the software.

[0014] Also, an eighth aspect of the present invention is the method according to the fifth aspect, wherein the third requirement includes an instruction to generate at least any one of the parameters, the endpoint, and the response body using column names included in the database design of the software.

[0015] Also, a ninth aspect of the present invention is the method according to the seventh or eighth aspect, further including a step of making a fourth request to a fourth generation AI model to generate a description of a database design using the one or more use cases before the second requirement and the third requirement.

[0016] Also, a tenth aspect of the present invention is the method according to any one of the first to ninth aspects, further including a step of transmitting at least any one of the generated one or more business logics and an API document corresponding to the business logic to the user terminal, and a step of receiving a modification to the business logic from the user terminal.

[0017] Moreover, an eleventh aspect of the present invention is the method of the tenth aspect, further including a step of making a fifth request to a fifth generative AI model to correct a use case in which the business logic can be used in accordance with the correction.

[0018] Moreover, a twelfth aspect of the present invention is the method of the eleventh aspect, further including a step of making a sixth request to a sixth generative AI model to correct API documentation corresponding to the business logic in accordance with the correction.

[0019] Moreover, a thirteenth aspect of the present invention is a method for creating software technical documentation, including steps of obtaining one or more use cases occurring in the software, making a request k to a generative AI model (k) to generate one or more business logics corresponding to the use case for at least any one of the one or more use cases, and making a request l to a generative AI model (l) to generate API documentation corresponding to the business logic for at least any one of the one or more business logics.

[0020] Moreover, a fourteenth aspect of the present invention is a program for causing a computer to execute a method for creating software technical documentation, the method including steps of obtaining one or more use cases occurring in the software, making a request k to a generative AI model (k) to generate one or more business logics corresponding to the use case for at least any one of the one or more use cases, and making a request l to a generative AI model (l) to generate API documentation corresponding to the business logic for at least any one of the one or more business logics.

[0021] Further, a 15th aspect of the present invention is an apparatus for creating software technical documents, which acquires one or more use cases occurring in the software, and for at least any one of the one or more use cases, performs a request k to request a generation AI model (k) to generate one or more business logics corresponding to the use case, and for at least any one of the one or more business logics, performs a request l to request a generation AI model (l) to generate API documents corresponding to the business logic.

Effect of the Invention

[0022] According to one aspect of the present invention, based on at least any one of one or more use cases obtained from explanatory data in which software explanations are expressed, one or more business logics available for the use case are generated, and by automating the process of further generating API documents corresponding to at least any one of the business logics by applying one or more generation AI models, the creation of technical documents required for software development can be significantly streamlined.

Brief Description of the Drawings

[0023]

Figure 1

Figure 2

Figure 3A

Figure 3B

Figure 3C

Figure 3D

Figure 3E

Figure 3F

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10A

Figure 10B

Figure 10C

Best Mode for Carrying Out the Invention

[0024] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0025] (First Embodiment) FIG. 1 shows an apparatus according to a first embodiment of the present invention. The apparatus 100 communicates via an IP network such as the Internet with a user terminal 110 used by a user and a platform 120 that provides a generation AI model in order to create technical documents necessary for software development. The generation AI model will be exemplarily described as being provided by a platform 120 capable of communicating with the apparatus 100, but an application for providing the generation AI model may be executed on the apparatus 100 so that the apparatus 100 provides the generation AI model.

[0026] The apparatus 100 includes a communication unit 101 such as a communication interface, a processing unit 102 such as a processor or a CPU, and a storage unit 103 including a storage device or a storage medium such as a memory or a hard disk, and can be configured by executing a program for performing each process or each operation in the processing unit 102. The apparatus 100 may include one or more apparatuses, computers or servers. Further, the program may include one or more programs, and may be recorded on a computer-readable storage medium to form a non-transitory program product. The program may be stored in a storage device or a storage medium such as a database 104 accessible via an IP network from the storage unit 103 or the apparatus 100, and instructions included in the program can be executed in at least one processor of the processing unit 102. Data described hereinafter as being stored in the storage unit 103 may be stored in a storage device or a storage medium such as the database 104, and vice versa.

[0027] First, the apparatus 100 receives description data (S201) in which a description of the software to be developed is expressed. The description data includes at least one of one or more sentences and one or more images. The one or more sentences may include one or more sentences describing the purpose of use, operation procedures, etc. of the software. Here, the description data is described as being received by the apparatus 100, but as long as the apparatus 100 can obtain the description data as a result. As an example, the apparatus 100 may receive a video about the software to be developed and analyze it to obtain the description data.

[0028] Next, based on the description data, the apparatus 100 generates a natural language description of a plurality of use cases that occur in the software. Hereinafter, mainly as an example, the case where the apparatus 100 receives a plurality of images in which a series of screens as shown in FIGS. 3A to 3F are expressed as the description data will be described. In this case, the apparatus 100 can generate a use case with each screen as the UI. Examples of the images included in the description data include, in addition to such GUIs, sequence diagrams showing the behavior of the software.

[0029] More specifically, as an example, use case generation can be performed by the apparatus 100 requesting (also referred to as the "first request") a generation AI model (also referred to as the "first generation AI model") to generate a plurality of use cases based on the description data (S202), the first generation AI model generating a plurality of use cases (S203), and transmitting the generated plurality of use cases to the apparatus 100 (S204). FIG. 2 shows an example in which a plurality of use cases are generated, but depending on the content of the description data, only one use case may be generated. Also, when the description data itself describes one or more use cases in natural language, the process of use case generation may not be performed.

[0030] Before making the first request, the device 100 may determine one or more data formats of the acquired description data and include an instruction in the first request according to the determination result for the first generation AI model. The said instruction is also called a prompt, and using this as an input, the first generation AI model generates an output. Specifically, when the description data is in text format, in image format, or includes both, different prompts may be included in the first request.

[0031] FIG. 4 shows an example of a use case generated by the first AI model according to the first embodiment of the present invention. The generated use case can be viewed by transmitting use case viewing screen display information, or more briefly, use case display information, from the device 100 to the user terminal 110 as, for example, an HTML-formatted file, and viewing it from the use case viewing screen 400 displayed on the display screen of the user terminal 110 on a web browser, or by operating an application (also called an "app") installed on the user terminal 110 and viewing it from the use case viewing screen 400 displayed on the said app using the received use case display information. Also, the output of the generated use case may be made on the command prompt on the user terminal 110. Note that the "viewing screen" can adopt various forms such as a web page, a modal window, a pop-up window, etc. when displayed on a web browser, and can be one screen of the app when displayed on the app. In any case, as long as it is a screen including an area for viewably displaying the generated use case, it corresponds to the viewing screen. The same technology can be applied to other screens mentioned in this specification. Although not shown here, the use case viewing screen 400 can enable the use case to be editable.

[0032] The use case displayed on screen 400 shown in FIG. 4 is a natural language description of the behavior of the attendance management system that provides the UIs shown in FIGS. 3A to 3F at the time of check-in, and corresponds to the UI shown in FIG. 3C. The use case includes, as items, an actor and a flow, and may further include at least one of a precondition and a postcondition. The use case 400 shown in FIG. 4 has an actor who is an employee, and as a precondition, the employee needs to be logged in to the attendance management system, and as a postcondition, it is determined that the employee's attendance is recorded on the current date. And the use case 400 describes a flow, that is, a series of behaviors of the system. Specifically, when the employee clicks the "Check In" button, the system is described as performing verification of the employee's identity, recording of attendance, and display of a confirmation message, and displaying an error message if the verification fails. Also, in the use case 400, three alternative flows are described. Here, when the use case at the time of check-in is described, the first AI model can generate use cases corresponding to each of the UIs shown in FIGS. 3A to 3F.

[0033] Next, for each of the generated multiple use cases, the apparatus 100 generates one or more business logics available for the use case. Here, "business logic" refers to the logic for processing data transmitted or received in communication with the UI based on business requirements. Here, "logic" means a series of instructions.

[0034] Business logic generation, more specifically, as an example, the apparatus 100 requests (also referred to as the "second request") a generation AI model (also referred to as the "second generation AI model") to generate one or more business logics available for the use case based on the use case (S205). The second generation AI model generates one or more business logics (S206) and transmits the generated one or more business logics to the apparatus 100 (S207). Although the above process has been described for each of the plurality of generated use cases, as in the example shown in FIG. 2, there may be a case where one or more business logics are generated for at least any one of the plurality of generated use cases.

[0035] FIG. 5 shows an example of business logic generated by the second generation AI model according to the first embodiment of the present invention. The browsing of the generated business logic can be performed on a business logic browsing screen 500 displayed on the display screen of the user terminal 110 by transmitting business logic browsing screen display information, or more briefly, business logic display information, from the apparatus 100 to the user terminal 110 in the same manner as the use case. Although not shown here, the business logic browsing screen 500 can be made editable for the business logic.

[0036] The business logic shown in the screen 500 of FIG. 5 corresponds to the use case shown in FIG. 4. The business logic may include, as one item, a logic that is a series of instructions and may further explicitly include at least one of the input and response items.

[0037] FIG. 6 shows an example of an instruction called a prompt that is included as part of the code for a second request for a second generation AI model according to a first embodiment of the present invention. In the prompt, as context, a use case in which one or more business logics to be generated can be utilized is set in the variable {use_case}. Further, the prompt can set, as context, a template (see FIG. 7) of one or more business logics to be generated in the variable {response_example}. The template may include, as one item, a logic that is a series of instructions, and may further explicitly include at least one of an input item and a response item. The prompt can set a database design in the variable {database_design} as context, which will be described in detail in the second embodiment.

[0038] The example of FIG. 6 is an example of a prompt described in natural language that is included as part of the code written in the programming language TypeScript. By executing the code, the OpenAI API can be called to generate one or more business logics available for the use case set in the second generation AI model provided on the platform 120. The OpenAI API is an example, and other APIs may be used. The programming language TypeScript is an example, and other languages may be used. The code for making the second request may be stored in the storage unit 103, and the device 100 may acquire this and execute the code obtained by setting the required values in the variables included in the code. Although a specific example of the first request for the first generation AI model is not shown, based on the description of the second request for the second generation AI model, those skilled in the art should be able to fully understand the implementation of the first request.

[0039] In FIG. 2, the first generation AI model and the second generation AI model are distinguished, but they may also be the same generation AI model. As the generation AI model, an LLM to which a transformer architecture is particularly applied is preferable, but it is assumed that the name of the architecture will change due to technological progress. Therefore, in this specification, the "transformer architecture" includes an architecture using one or more features of the transformer architecture or its improvement. Whether the "generation AI model" in this specification is the same is determined by whether the types of the generation AI models specified by the user are the same. For example, in the case of the OpenAI API, if the value of the variable "model" is the same, it is expressed as being the same as the generation AI model. When the first generation AI model and the second generation AI model are not the same, they may be provided on the same platform 120. In the example of FIG. 2, where a second request is made after the first request, if the first generation AI model and the second generation AI model are the same generation AI model or generation AI models provided on the same platform, these requests may be made by a single call to the API. Also, needless to say, the first and second requests may each include a plurality of requests and may include one or more processes performed by the device 100 other than the requests for the generation AI model.

[0040] If the set use case is long and the length of the instruction for the second generation AI model included in the second request exceeds the upper limit set for the second generation AI model, a summary generation of the use case may be requested from a generation AI model provided on the platform 120 or another platform, and the obtained summary may be set in the variable {use_case} and the second request may be made.

[0041] Next, for each of the generated one or more business logics, the device 100 generates an API document corresponding to each business logic. Here, the "API document" refers to a document describing the functions and usage methods of the API.

[0042] API document generation, more specifically, as an example, the apparatus 100 requests (also referred to as the "third request") the generation AI model (also referred to as the "third generation AI model") to generate an API document corresponding to each of the generated one or more business logics (S208), the third generation AI model generates an API document (S209), and the generated API document can be transmitted to the apparatus 100 (S210). Although the case where the above processing is performed for each of the generated one or more business logics has been described as in the example shown in FIG. 2, there may be a case where an API document is generated for at least any one of the generated multiple business logics.

[0043] FIG. 8 shows an example of an API document according to the first embodiment of the present invention. The API document displayed on the screen 800 in FIG. 8 includes, as items, a request type, a parameter, an endpoint, an API response code, and a response body, and may further include at least any one of validation, authentication, authorization, and a request body. Although not shown in FIG. 8 due to space limitations, the API document preferably includes one or more business logics corresponding to the API document, as shown in FIG. 10C, so that it is possible to grasp for which implementation of which business logic the API described in the API document can be used. One API may be available for multiple business logics.

[0044] The viewing of the API document can be performed, for example, on the API document viewing screen 800 displayed on the display screen of the user terminal 110 by transmitting the API document viewing screen display information, or more briefly, the API document display information, from the device 100 to the user terminal 110 in the same way as the use case. At this time, it is preferable to describe the API document display information in the response body in a JSON format so that it can be displayed. Although not shown here, the API document viewing screen 800 can be made editable for the API document.

[0045] FIG. 9 shows an example of a prompt described in natural language that is partially included for a third request for the third generation AI model according to the first embodiment of the present invention. In the prompt, as the context, one or more business logics corresponding to the generated API document can be set in the variable {business_logics}. Further, in the prompt, as the context, a template of the generated API document can be set in the variable {response_example}. The items that such a template may include are as described above. In the example of FIG. 9, in addition to the business logic, values are set as variables in the context for the use cases in which the business logic can be used. However, if the business logic is given, it is possible to cause the third generation AI model to generate the corresponding API document. In the prompt, as the context, the database design can be further set in the variable {database_design}, which will be described in the second embodiment.

[0046] The example in FIG. 9 is a part included in the code written in the programming language TypeScript. By executing the code, the OpenAI API can be called to generate one or more API documents corresponding to the business logic set in the third generation AI model provided on the platform 120. The OpenAI API is an example, and other APIs may be used. The programming language TypeScript is an example, and other languages may be used. The code for making the third request may be stored in the storage unit 103, and the device 100 can obtain it and execute the code obtained by setting the required values in the variables included in the code.

[0047] In FIG. 2, the case where the third generation AI model is the same as the second generation AI model is shown, but they may be different generation AI models. When the second generation AI model and the third generation AI model are not the same, they may be provided on the same platform 120. In the example of FIG. 2, a third request is made after the second request. When the second generation AI model and the third generation AI model are the same generation AI model or generation AI models provided on the same platform, these requests may be made by a single call to the API. Also, the third request may include multiple requests and may include one or more processes performed by the device 100 other than the request for the third generation AI model.

[0048] The device 100 may receive the generated one or more use cases and the technical documents generated using them and transmit them to the user terminal 110 (S211). Here, specifically, the generated one or more business logics and the one or more API documents corresponding to them are included in the technical documents.

[0049] As described above, based on at least any one of one or more use cases obtained from the description data in which the description of the software is expressed, one or more business logics available for the use case are generated, and the process of further generating API documents corresponding to at least any one of the business logics is automated by applying one or more generation AI models, whereby the creation of technical documents required for software development can be significantly streamlined.

[0050] Before proceeding to the generation of business logic using use cases, verification of the use cases may be performed. The verification can be, for example, verification that the use case satisfies the conditions necessary for generating business logic using it. Also, before proceeding to the creation of API documents using business logic, verification of the business logic may be similarly performed.

[0051] (Second Embodiment) As in the examples of the prompts in FIGS. 6 and 9 described in the first embodiment, as context, a description of database design may be set in the variable {database_design}. In this case, as in the example of FIG. 6, an instruction to determine the input items included in the business logic using the column names included in the database design can be included in the second request. By doing so, business logic that is consistent with the database design can be obtained. Also, as in the example of FIG. 9, an instruction to determine at least any one of the items of parameters, endpoints, verification, request body, and response body included in the API document using the column names included in the database design can be included in the third request.

[0052] The description of the database design may be obtained by the apparatus 100 receiving it from the user terminal 110, or may be generated by a generative AI model based on one or more use cases generated or obtained by the apparatus 100. More specifically, the apparatus 100 may request (also referred to as a "fourth request") a generative AI model (also referred to as the "fourth generative AI model") to generate a database design based on one or more use cases. Before proceeding with the generation of the database design based on one or more use cases, the verification of the one or more use cases may be performed. The verification may be, for example, the verification that the one or more use cases satisfy the conditions necessary for generating the database design based thereon. Also, the fourth request may include a plurality of requests and may include one or more processes performed by the apparatus 100 other than the request for the fourth generative AI model.

[0053] Although a specific example of the fourth request for the fourth generative AI model is not shown, based on the descriptions of the second request for the second generative AI model and the third request for the third generative AI model, a person skilled in the art should be able to fully understand the implementation of the fourth request. The fourth request, like the second and third requests, can be made possible by preparing the code in advance and having the apparatus 100 set the necessary variables and execute it. The fourth generative AI model may be the same as at least one of the first to third generative AI models or provided on the same platform or apparatus. Of course, if the variable {database_design} is set in the second and third requests, it is necessary to execute the fourth request before the second and third requests.

[0054] (Third Embodiment) The technical document sent to the user terminal 110 can be displayed in an editable format on the display screen of the user terminal 110, and the user of the user terminal 110 can make modifications (S212). Then, the added modifications are sent from the user terminal 110 to the apparatus 100, and the apparatus 100 can request (also referred to as the "fifth request") the generative AI model (also referred to as the "fifth generative AI model") to modify at least any one of the use cases according to the modification. More specifically, when the user makes a modification to a certain business logic, the fifth request can be a request to modify the use cases that utilize the business logic (S213).

[0055] Generally, since use cases are created by members on the business side and technical documents are created by members on the development side, it is not easy to maintain the consistency between use cases and the technical documents based on them. By performing the modification of the use cases according to the present embodiment, it becomes realistically possible to maintain the consistency between the two.

[0056] When a modification to the business logic is made, in addition to or instead of modifying the use cases, it is further possible to request (also referred to as the "sixth request") the generative AI model (also referred to as the "sixth generative AI model") to modify the API document corresponding to the business logic.

[0057] Similar to the second request and the third request, the fifth request and the sixth request can be made possible by preparing the code in advance and having the apparatus 100 set the necessary variables and execute it. The fifth generative AI model and the sixth generative AI model may be the same as at least any one of the first to third generative AI models or provided on the same platform or apparatus. Also, the fifth and sixth requests may each include a plurality of requests and may include one or a plurality of processes performed by the apparatus 100 other than the requests to the generative AI model.

[0058] Note that in the above-described embodiments, if there is no description of "only", such as "based only on", "responding only to", "only in the case of", "referring only to", etc., it should be noted that additional information may be considered in this specification. Also, as an example, the description "do b in the case of a" does not necessarily mean "always do b in the case of a" or "do b immediately after a", except when explicitly stated. Also, the description "each a that constitutes A" does not necessarily mean that A is constituted by a plurality of components, and includes the case where the component is singular.

[0059] Also, just in case, even if there is an aspect of performing an operation different from the operations described in this specification in any method, program, terminal, device, server, or system (hereinafter referred to as "method, etc."), each aspect of the present invention is directed to the same operation as any of the operations described in this specification, and it is added that the existence of an operation different from the operations described in this specification does not exclude the method, etc. from the scope of each aspect of the present invention.

[0060] Also, in the above description, when referring to a plurality of generation AI models, for example, in the second embodiment, it may be necessary to request the fourth AI model before requesting the second generation AI model and the third generation AI model. In such a case, from the viewpoint of readability, the fourth AI model may be called "AI model (k)", the first AI model may be called "AI model (l)", the second AI model may be called "AI model (m)", and the third AI model may be called "AI model (n)". For other cases, it may be similarly readapted as appropriate.

Explanation of Reference Numerals

[0061] 100 Device 101 Communication Unit 102 Processing Unit 103 Storage Unit 104 Database 110 User Terminal 120 Platform 400 Use Case Browsing Screen 500 Business Logic Browsing Screen 800 API Document Browsing Screen

Claims

1. A method for creating a technical document of software, comprising: obtaining description data in which the description of the software is expressed, the description data including at least one of one or more sentences and one or more images; making a first request to a first generation AI model to generate one or more use cases that occur in the software based on the description data; making a second request to a second generation AI model to generate one or more business logics corresponding to at least one of the one or more use cases; making a third request to a third generation AI model to generate API documentation corresponding to at least one of the one or more business logics; and.

2. The method according to claim 1, wherein the description data includes one or more images.

3. The method according to claim 1 or 2, wherein the second request includes a template of a business logic having items of input, logic, and response.

4. The method according to claim 1 or 2, wherein the third request includes a template of API documentation having items of request type, parameter, endpoint, API response code, and response body.

5. The method according to claim 3, wherein the second request includes an instruction to generate the input item using a column name included in the database design of the software.

6. The method according to claim 5, wherein Before the fourth claim, further including the step of making a fourth claim that requires a fourth generation AI model to generate a natural language description of database design using the one or more use cases.

7. The method according to claim 1 or 2, transmitting at least any one of the generated one or more business logics and API documents corresponding to the business logic to the user terminal; receiving a modification to the business logic from the user terminal and further including.

8. A program for causing a computer to execute a method for creating technical documents of software, the method comprising: obtaining description data in which the description of the software is expressed, the description data including at least one of one or more sentences and one or more images; making a first request that requires a first generation AI model to generate one or more use cases occurring in the software based on the description data; making a second request that requires a second generation AI model to generate one or more business logics corresponding to at least any one of the one or more use cases; making a third request that requires a third generation AI model to generate API documents corresponding to at least any one of the one or more business logics and including.

9. An apparatus for creating technical documents of software, obtaining description data in which the description of the software is expressed, the description data including at least one of one or more sentences and one or more images, Make a first request to a first generation AI model to generate one or more use cases that occur in the software based on the description data. Make a second request to a second generation AI model to generate one or more business logics corresponding to at least one of the one or more use cases. Make a third request to a third generation AI model to generate API documents corresponding to at least one of the one or more business logics.

Images