Task processing method and device based on AI large model and storage medium

Abstract

The invention discloses a task processing method and device based on an AI large model and a storage medium, and relates to the technical field of artificial intelligence, the disclosed task processing method based on the AI large model comprises the steps that in a display interface, a dialogue picture of an intelligent assistant and an engineering picture of a target integrated development environment are displayed, the dialogue picture comprises a first area and a second area; in response to a sending operation of the development demand description information in the first area, displaying a development task obtained by planning of the AI large model based on context information associated with the development demand description information and the target project instance in a second area; and in response to a trigger operation of a control associated with the development task, displaying an execution result of the development task on the engineering picture, and displaying a response result of the execution result in the second area. The method aims at improving the development efficiency of project engineering and shortening the development period of the project engineering.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to task processing methods, devices and storage media based on large AI models. Background Technology

[0002] Currently, an Integrated Development Environment (IDE) is an application that provides a program development environment. Most mainstream IDEs now offer basic development tasks such as code editing, debugging, and version control, providing developers with a good development environment. However, project development still relies heavily on manual processes. As projects become larger and more complex, manual development leads to longer development cycles. Summary of the Invention

[0003] The main purpose of this application is to provide a task processing method, device and storage medium based on a large AI model, which aims to improve the development efficiency of project engineering and shorten the development cycle of project engineering.

[0004] To achieve the above objectives, this application proposes a task processing method based on a large AI model, including: The display interface shows the dialogue screen of the intelligent assistant and the engineering screen of the target integrated development environment. The dialogue screen includes a first area and a second area. In response to the sending of development requirement description information in the first area, the AI ​​big model is displayed in the second area to plan the development tasks based on the context information associated with the development requirement description information and the target project instance. Responding to the triggering operation of the control associated with the development task, the execution result of the development task is displayed on the project screen, and the response result of the execution result is displayed in the second area.

[0005] In one embodiment, in response to the sending operation of development requirement description information input in the first area, the AI ​​big model is displayed in the second area based on the context information associated with the development requirement description information and the target project instance, and the planned development tasks include: In response to the sending operation of the development requirement description information entered in the first area, obtain the context information associated with the target project instance; Integrate development requirement description information and contextual information to construct the initial prompt message; The first prompt is sent to the AI ​​model, which then uses the first prompt to infer the development task. The second area displays the development tasks obtained from the plan.

[0006] In one embodiment, responding to a trigger operation of a control associated with a development task, displaying the execution result of the development task on the project screen, and displaying the response result of the execution result in a second area includes: Respond to the triggering operation of the control associated with the development task and obtain the execution tool corresponding to the development task in the integrated development environment; Invoke the execution tool to perform development tasks; The execution results of the development task are displayed on the engineering screen, and the response results of the execution results are displayed in the second area.

[0007] In one embodiment, the response results displayed in the second region include: If the execution tool call fails, the second area displays the first response result; or, the file acquisition tool is called to obtain the log file, the log file is parsed, and the first response result is obtained; the first response result is then displayed in the second area. If the execution tool call is successful, the second response result will be displayed in the second area; The response results include either the first response result or the second response result.

[0008] In one embodiment, after invoking the execution tool to perform the development task, the method further includes: If the execution tool call fails, the response result, context information, development task, and execution result of the development task are integrated to obtain a second prompt message; The second prompt is sent to the AI ​​model, which then uses the second prompt to make inferences and adjust the development task. The adjusted development tasks are displayed in the second area; In response to the triggering operation of the control associated with the adjusted development task, the execution result of the adjusted development task is displayed on the project screen, and the response result of the execution result of the adjusted development task is displayed in the second area.

[0009] In one embodiment, after responding to a trigger operation of a control associated with a development task, displaying the execution result of the development task on the engineering screen, and displaying the response result of the execution result in a second area, the method further includes: If there are unexecuted development tasks, the unexecuted development tasks are corrected based on the execution results of the executed development tasks and the context information associated with the target project instance. In response to the triggering operation of the control associated with the corrected unexecuted development task, the execution result of the corrected unexecuted development task is displayed on the project screen, and the response result of the execution result of the corrected unexecuted development task is displayed in the second area.

[0010] In one embodiment, after responding to a trigger operation of a control associated with a development task, displaying the execution result of the development task on the engineering screen, and displaying the response result of the execution result in a second area, the method further includes: If all development tasks are completed, the email tool is invoked based on the development requirement description information; the target email is generated and pushed according to the execution results and the response results.

[0011] In one embodiment, the task processing method based on a large AI model further includes: In response to the scanning operation of integrated development environment instances, all scanned integrated development environment instances are displayed in the second area; In response to a triggered operation targeting any one of the said integrated development environment instances, the target integrated development environment instance is determined.

[0012] In addition, to achieve the above objectives, this application also proposes a task processing device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the task processing method based on the AI ​​large model as described above.

[0013] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the task processing method based on the AI ​​large model described above.

[0014] This application displays a dialogue screen of an intelligent assistant and a project screen of the target integrated development environment on a display interface. By operating on the visualized screen, users can complete a series of operations, including sending development requirement description information, planning development tasks based on the development requirement description information and context information associated with the target project instance, generating execution results of development tasks, and providing feedback on the execution results. Among these operations, the development task planning and processing are automatically performed by an AI large model, and the development tasks are automatically processed by the integrated development environment. Compared with related technologies where tasks are manually processed in project engineering, this improves the efficiency of project engineering task processing and shortens the project engineering development cycle. Attached Figure Description

[0015] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating an embodiment of the task processing method based on a large AI model in this application. Figure 2 This is a schematic diagram illustrating the interaction between the multi-IDE terminal, intelligent assistant, and large AI model in this application; Figure 3 This is a system architecture diagram of the task processing method based on a large AI model in this application; Figure 4 This is a data flow timing diagram of the task processing method based on a large AI model in an embodiment of this application.

[0018] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0020] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0021] Currently, an Integrated Development Environment (IDE) is an application that provides a program development environment. Most mainstream IDEs now offer basic development tasks such as code editing, debugging, and version control, providing developers with a good development environment. However, project development still relies heavily on manual processes. As projects become larger and more complex, manual development leads to longer development cycles.

[0022] To address the aforementioned issues, this application proposes a task processing method based on an AI large model. The main technical solution includes: displaying a dialogue screen of an intelligent assistant and a project screen of the target integrated development environment in a display interface, wherein the dialogue screen includes a first area and a second area; responding to the sending operation of development requirement description information in the first area, displaying the development task planned by the AI ​​large model based on the context information associated with the development requirement description information and the target project instance in the second area; responding to the triggering operation of the control associated with the development task, displaying the execution result of the development task in the project screen, and displaying the response result of the execution result in the second area.

[0023] This application displays a dialogue screen of an intelligent assistant and a project screen of the target integrated development environment on a display interface. By operating on the visualized screen, users can complete a series of operations, including sending development requirement descriptions, planning development tasks based on the context information associated with the development requirement descriptions and the target project instance, generating the execution results of the development tasks, and providing feedback on the execution results. Among these operations, the development task planning and processing are automatically performed by an AI large model, and the development tasks are automatically processed by the integrated development environment. Compared with related technologies where tasks are manually processed by hand, this improves the efficiency of task processing and shortens the development cycle of the project.

[0024] Furthermore, during the development task planning process, this application uses the context information associated with the target project instance as input for development task planning, which enables the association of development requirements with the actual control scenario of the project. This makes the generated development tasks more consistent with the actual control scenario of the project, improves the accuracy of the planned development tasks, and thus improves the accuracy of the execution results of the development tasks.

[0025] It should be noted that the executing entity in this embodiment can be a computing service device with task processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device or task processing device capable of performing the above functions. The following description uses a task processing device as an example to illustrate this embodiment and the subsequent embodiments.

[0026] It should be noted that this application also provides a task processing system, which includes the aforementioned task processing device, on which an IDE and a smart assistant run. The smart assistant can be integrated into the IDE as a plugin, and the display interface of the task processing device can show the dialogue screen of the smart assistant and the corresponding project screen of the IDE. Alternatively, the smart assistant can exist independently of the IDE. The IDE and the smart assistant will be described in detail below: An IDE can be an IDE for a PLC (Programmable Logic Controller). It can also be other types of IDEs, such as those with code editing, debugging, and version control functions, all of which are included in this application. This application uses a PLC IDE as an example, which can be used for project development in industrial equipment-related fields. This IDE can receive development tasks planned by an intelligent assistant, execute the corresponding development tasks sequentially in the project screen, display the execution results of the development tasks, and feed back the response results of the development task execution to the intelligent assistant.

[0027] The intelligent assistant acts as an information transmission bridge between the IDE and the AI ​​big model. It is used to receive development requirement description information, call the AI ​​big model to plan development tasks based on the context information associated with the development requirement description information and the target project instance, display the planned development tasks, send the planned development tasks to the IDE for processing in sequence, and receive the response results fed back by the IDE after processing the development tasks.

[0028] The aforementioned large-scale AI model, such as GPT-4, possesses natural language understanding and code generation capabilities. It is used for tasks such as planning development tasks based on development requirements and contextual information, and for correcting the planned tasks. This large-scale AI model can be pre-trained using knowledge relevant to the field of this application, or it can be a general-purpose large-scale AI model, or it can be obtained through fine-tuning.

[0029] The aforementioned IDE and smart assistant communicate via the MCP protocol.

[0030] Based on this, the embodiments of this application provide a task processing method based on a large AI model, referring to... Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the task processing method based on a large AI model according to this application.

[0031] In this embodiment, the task processing method based on the AI ​​large model includes steps S10 to S30: Step S10: In the display interface, the dialogue screen of the intelligent assistant and the engineering screen of the target integrated development environment are displayed, wherein the dialogue screen includes a first area and a second area.

[0032] The display interface refers to a visual window that users can operate intuitively. It is a unified carrier integrating dialog boxes and project screens, and can be the main interface of the software. This display interface can also be the graphical user interface of a task processing device.

[0033] The dialog box is a functional area for user interaction with the AI ​​model and the IDE. It is divided into two areas, the first and the second. Their core functions are to convey development requirements, display AI feedback, and show the integrated development environment's (IDE) response results. The first area is the user input field where users can enter text and upload files to describe their development needs. The second area displays the content generated by the AI ​​model and the IDE's feedback results, specifically including planned development tasks and the response results of their execution. The specific locations of the first and second areas are not limited; for example, the first area can be located at the bottom of the dialog box, and the second area can be located at the top.

[0034] An Integrated Development Environment (IDE) is a specialized software tool used for software development, providing one-stop functionality for code writing, compilation, and debugging. When a smart assistant is integrated into an IDE, the target IDE can be the IDE that integrates the smart assistant. When the smart assistant exists independently of the IDE, the target IDE can be any IDE detected by the smart assistant, and one of the detected IDEs can be selected as the target IDE as needed.

[0035] Among them, the target project instance refers to the collection of project files that the user is developing in the current target integrated development environment, including code files, configuration files, dependency libraries, etc., which is the core basis for AI analysis.

[0036] The project screen, located within the IDE, is an area used to display the actual running effects of development tasks, such as code compilation results, program running logs, interface rendering effects, device connection relationships, and code.

[0037] In one feasible approach, a column layout technique can be used, which can achieve horizontal or vertical column layout.

[0038] Example 1: When the display interface is the main interface of the IDE, window splitting controls such as the split view function can be used to display the target project instance's project screen on the left and the dialog screen on the right; alternatively, the target project instance's project screen can be displayed on the right and the dialog screen on the left; the target project instance's project screen can be displayed at the top of the display interface and the dialog screen at the bottom; or the target project instance's project screen can be displayed at the bottom of the display interface and the dialog screen at the top. After using the column layout technology, the column ratio can be adjusted by dragging according to the actual situation, improving the display effect of the dialog screen and the project screen. By displaying the intelligent assistant's dialog screen and the target integrated development environment's project screen simultaneously in the same display interface, user operation jumps are reduced.

[0039] Example 2: When the display interface is a graphical user interface for a task processing device, the interface can use window splitting controls such as the split view function to display the target project instance's project screen on the left and the dialog screen on the right; alternatively, the target project instance's project screen can be displayed on the right and the dialog screen on the left; the target project instance's project screen can be displayed at the top of the display interface and the dialog screen at the bottom; or the target project instance's project screen can be displayed at the bottom of the display interface and the dialog screen at the top. After using the column layout technology, the column ratio can be adjusted by dragging according to the actual situation, improving the display effect of the dialog screen and the project screen. By simultaneously displaying the intelligent assistant's dialog screen and the target integrated development environment's project screen on the same display interface, the user's operation jump is reduced.

[0040] Another feasible approach is to use pop-up embedding technology. The dialog window can be designed as a floating pop-up that can be minimized or maximized without obscuring the project view, such as the IDE's code editing area or console. Users can switch between interactive and execution perspectives at any time. Specifically, a shortcut control for the dialog window can be embedded in the IDE. Triggering this shortcut control will bring up the dialog window. This method allows the dialog window to be displayed only when needed and hidden when not in use, preventing it from obscuring the project view.

[0041] The above method is applicable to application scenarios where intelligent assistants are integrated into the IDE.

[0042] If the intelligent assistant exists independently of the IDE, it can display the dialog box and the project view independently in the display interface, with the dialog box display not dependent on the project view. During the display process, the display ratio of the dialog box and the project view can be adaptively adjusted in the display interface to improve the display effect.

[0043] Step S20: In response to the sending operation of the development requirement description information in the first area, the AI ​​big model is displayed in the second area to plan the development tasks based on the context information associated with the development requirement description information and the target project instance.

[0044] The development requirement description information refers to the content entered by the user in the first area to express the development goals. This can be text descriptions, code snippets, error logs, etc., and the description format is not limited to text; it can also include voice, images, and other descriptive formats. For example, when describing the development requirement information in text form, it could be something like: "Write a program for 6 axes," "Help me interpret the current project," "Help me check the last error log and explain these error messages," or "Create a project where the PLC device uses AC81X-0122-U0R0, based on the EtherCAT bus, and the project contains two SV660_1Axis_00915s, one axis rotating 360 degrees per second, and the other moving linearly at 20 millimeters per second." This development requirement description information can be historical or pre-set templates. It is encapsulated as controls and displayed in the first area of ​​the dialog box for user selection. When one of the development requirement description controls is triggered, the associated development requirement description information is sent, thus completing the action of sending the development requirement description information in the first area. Alternatively, these development requirement descriptions can be edited in the first area. In this case, the first area includes an edit box and a send control. After entering the development requirement description in the edit box, the edited development requirement description is sent by triggering the send control, thus completing the action of sending the development requirement description in the first area.

[0045] Contextual information refers to auxiliary information related to the target engineering instance, which is dynamically updated as the task is processed. Different target engineering instances are associated with different contextual information. This contextual information facilitates the AI ​​model's understanding of the actual engineering control scenario, enabling the AI ​​model to correlate development requirements with the actual engineering control scenario during inference. This makes the generated development tasks more consistent with the actual engineering control scenario, improving the accuracy of the planned development tasks and ultimately enhancing the accuracy of the execution results. Specifically, contextual information includes, but is not limited to, basic information of the integrated development environment (IDE), equipment information associated with the target engineering instance, and programming information associated with the target engineering instance. Taking a PLC IDE as an example, basic IDE information includes, but is not limited to: the IDE version, currently open windows, the directory location of project files and logs, and communication ports. Equipment information associated with the target engineering instance includes, but is not limited to, the attribute information of the engineering equipment used by the target engineering instance, such as the equipment's model, size, and power. Programming information associated with the target engineering instance can be the PLC's execution logic, code execution logic, or graphical execution logic.

[0046] In this context, development tasks refer to the specific execution steps broken down by the AI ​​model based on development requirement descriptions and contextual information. For example, for the development requirement description "write a program for 6 axes," the planned development tasks include: project creation task, equipment configuration task, and program writing task. The types and number of development tasks derived from different development requirement descriptions vary.

[0047] In one feasible approach, in response to the sending operation of the development requirement description information in the first area, the system automatically retrieves the key context of the target project instance from the IDE, concatenates it with the development requirement description information to form a prompt, and sends it to the AI ​​big model. The AI ​​big model performs reasoning and analysis on the prompt information to obtain the development task corresponding to the development requirement description information, thereby achieving the purpose of planning the development task.

[0048] Step S30: Respond to the trigger operation of the control associated with the development task, display the execution result of the development task on the project screen, and display the response result of the execution result in the second area.

[0049] The second area displays development tasks and their associated controls. Triggering these controls confirms the receipt of the development tasks planned by the AI ​​model based on the development requirements description and the context information associated with the target project instance. Responding to the triggering of the control associated with the development task, the execution result of the development task is displayed on the project screen, and the response result is displayed in the second area. The second area can also display individual planned development tasks, each functioning as a control. Triggering the corresponding development task displays its execution result on the project screen and the response result is displayed in the second area.

[0050] The execution result is the actual output of the development task after it is run in the IDE. For example, for the development requirement description information "write a program with 6 axes", the corresponding execution result may be to create a new project and jump to display the display interface of the new project, display the added master station equipment, and display the generated program, including code program or ladder diagram, etc.

[0051] The response result is an interpretation of the execution result. If the execution result is normal, the response result may be "Execution successful: Project has been created" or "Execution successful: Main site creation completed". If the execution result fails, the response result may be "Tool call failed" and the reason for the failure.

[0052] In one feasible approach, the IDE is invoked to execute the development task in response to the triggering operation of the control associated with the development task. After the development task is executed, the execution result of the development task is displayed on the project screen. The IDE generates a response result of the execution result of the development task and feeds the response result back to the smart assistant, which displays the response result of the execution result in the second area of ​​the smart assistant.

[0053] In this embodiment, the intelligent assistant's dialogue screen and the target integrated development environment's engineering screen are displayed on the display interface. By operating on the visualized screen, a series of operations are completed, including sending development requirement description information, planning development tasks based on the development requirement description information and the context information associated with the target engineering instance, generating the execution results of the development tasks, and providing feedback on the execution results. Among these, the development task planning and processing are automatically performed by the AI ​​large model, and the development tasks are automatically processed by the integrated development environment. Compared with the manual processing of project engineering tasks by related technologies, this improves the efficiency of project engineering task processing and shortens the project engineering development cycle.

[0054] In other embodiments, other controls associated with the switch task may also be displayed in the second area. Triggering these other controls allows the AI ​​model to reject the development task planned based on the development requirement description information and the context information associated with the target project instance. When these other controls are triggered, the AI ​​model is re-invoked to plan the development task based on the development requirement description information and the context information associated with the target project instance.

[0055] In one feasible implementation, step S20 includes: Step S21: Respond to the sending operation of the development requirement description information entered in the first area and obtain the context information associated with the target project instance; In one feasible approach, in response to the sending operation of the development requirement description information input in the first area, the context information acquisition tool corresponding to the target project instance is invoked based on the listening parameters of the target project instance, and the context information associated with the target project instance is obtained through the context information acquisition tool.

[0056] In another feasible approach, API interfaces with the target project instance's data storage system, such as a database or file system, can be predefined within the system. After the development requirements description information is entered in the first area and the send operation is executed, the system retrieves the associated context information from the target project instance's storage location by calling the corresponding API interface.

[0057] In another feasible approach, a message queue can be constructed. When operations related to the target project instance occur, the resulting context information is sent as a message to the message queue. Upon responding to the sending operation of the development requirement description information entered in the first area, the system retrieves the latest message related to the target project instance from the message queue, thereby obtaining the context information. Because the message queue allows for timely acquisition of the latest context information of the target project instance, it ensures that the acquired information is synchronized with the actual situation, avoiding inaccurate development task planning due to information lag.

[0058] Step S22: Integrate development requirement description information and context information to construct the first prompt information; The first prompt information is a type of input used to convey specific task-related information to the AI ​​model. It integrates development requirement description information and contextual information associated with the target engineering instance, aiming to provide the AI ​​model with comprehensive and accurate information so that it can reason based on this information and generate development tasks that meet the requirements.

[0059] In one feasible approach, development requirement descriptions and contextual information can be concatenated according to a specific format. For example, a fixed template can be defined first, placing the development requirement descriptions in a specific location within the template, and the contextual information categorized into different sections within the template. Then, the template can be formatted to conform to the text format required for inputting large AI models.

[0060] In another feasible approach, information fusion can be achieved using natural language processing (NLP) techniques: NLP techniques such as information extraction and semantic understanding are used to deeply analyze the development requirements description and contextual information. Key information is extracted, the semantic relationships between them are understood, and then, based on these analysis results, the information is reorganized in natural language to construct a more accurate and clear initial prompt.

[0061] Step S23: The first prompt information is sent to the AI ​​big model, whereby the AI ​​big model performs inference based on the first prompt information to obtain the development task; In one feasible approach, the initial message can be encapsulated in the message body via the HTTP protocol, with necessary authentication information set in the message header to ensure the message's legitimacy and security. The message, consisting of the message body and header, is then sent to the API interface provided by the AI ​​large-scale model. Upon receiving the message through the API interface, the AI ​​large-scale model parses the message body, obtains the initial message, performs inference, and then returns the inference result.

[0062] In another feasible approach, the initial notification can be sent to a message broker such as RabbitMQ or Kafka. The AI ​​model, acting as a consumer, retrieves and processes the message from the broker. This method enables asynchronous communication, improving system responsiveness and scalability, especially when handling a large number of development requests, thus avoiding performance bottlenecks caused by synchronous communication.

[0063] Step S24: Display the planned development tasks in the second area.

[0064] In one feasible approach, common front-end frameworks can be used to bind development task data to corresponding components. Through the framework's rendering mechanism, detailed information about the development task, such as task name, task description, task priority, and the name of the execution tool, is displayed in an intuitive interface in the second area.

[0065] In another feasible approach, for complex development tasks, a chart library can be used to visualize the tasks in the form of charts. For example, Gantt charts can be used to display the task's timeline, and flowcharts can be used to display the task's execution flow, allowing developers to gain a clearer understanding of the overall task.

[0066] In this embodiment, by integrating development requirement description information and contextual information in a reasonable manner, it is ensured that the initial prompt accurately conveys the development requirements and contextual information, avoiding information loss or misunderstanding, and enabling the AI ​​model to perform reasoning based on accurate information. Furthermore, the constructed initial prompt is presented in a clear and easy-to-understand manner, helping the AI ​​model better understand the task background and requirements, improving the efficiency and accuracy of reasoning. By intuitively displaying the development task in the second area, developers can quickly understand the content and requirements of the task without spending a lot of time interpreting complex data or documents, improving the efficiency of information acquisition. The intuitive display also facilitates further operations by developers on the development task, improving the convenience and flexibility of development management.

[0067] In other embodiments, in response to the sending operation of development requirement description information input in the first area, the system can obtain context information associated with the target project instance and obtain reference examples; integrate the development requirement description information, context information, and reference examples to construct first prompt information; send the first prompt information to the AI ​​big model, whereby the AI ​​big model performs reasoning based on the first prompt information to obtain development tasks; and display the planned development tasks in the second area. The aforementioned reference examples guide the reasoning process of the AI ​​big model, and these reference examples include development task samples corresponding to different user intent samples. The AI ​​big model can parse the requirement description information and context information to determine the user intent, calculate the similarity between the parsed user intent and each user intent sample, determine the matching target user intent sample based on the similarity, and finally output and display the development task sample associated with the target user intent sample as the planned development task. Since the reference examples are determined by historical development projects, the AI ​​big model can refer to historical projects for development task planning, eliminating the need for cumbersome reasoning and improving the efficiency of development task planning.

[0068] In one feasible implementation, step S30 includes: Step S31: Respond to the trigger operation of the control associated with the development task and obtain the execution tool corresponding to the development task in the integrated development environment; In an integrated development environment (IDE), execution tools refer to the software or program components used to actually complete specific development tasks. Different development tasks require different execution tools. These can be code compilers, which convert high-level programming language code into machine-executable code (in practice, intelligent assistants can send code compilation instructions to the code compiler to convert high-level programming language code into machine-executable code); code debuggers, which help developers find and fix errors in the code; or version control tools, used to manage code version changes, etc. Different development tasks require different execution tools to complete.

[0069] In one feasible approach, specific attribute identifiers are pre-defined for each development task, and these attribute identifiers have a pre-defined mapping relationship with the execution tools. When a control associated with a development task is triggered, the system reads the attribute identifier of that development task, and then searches in the execution tool library according to the mapping relationship to obtain the corresponding execution tool.

[0070] In another possible approach, an event listener mechanism is used to capture trigger events of controls associated with development tasks. When a control of a development task is triggered, the system executes the corresponding event handler function. Within the event handler function, a pre-configured configuration file is read, which records the correspondence between different development tasks and execution tools. Based on the development task represented by the triggering control, the corresponding execution tool information is parsed from the configuration file, and thus the execution tool is obtained.

[0071] Because predefined mapping relationships or configuration files can accurately obtain the execution tool corresponding to the development task, errors that may occur due to manually selecting the execution tool are avoided, ensuring that the development task can be executed using the correct tool.

[0072] Step S32: Invoke the execution tool to perform the development task; In one feasible approach, if the execution tool provides an API interface, the intelligent assistant can use the corresponding programming language to call the interface of that execution tool in the IDE to perform development tasks. For example, some code analysis tools provide API interfaces; the system can send a request to this interface, passing relevant parameters of the development task, and the analysis results returned by the interface are the execution results of the development task.

[0073] Step S33: Display the execution results of the development task on the engineering screen and display the response results of the execution results in the second area.

[0074] In one feasible approach, during the execution of development tasks by the tool, the execution results are sent to the front end in real time via a WebSocket connection. Upon receiving the results, the front end displays them in real-time on the project screen. Simultaneously, the front end analyzes the execution results in real-time, obtains the response results, and sends the response results to a second area for display via a WebSocket connection. This method enables real-time updates of both execution and response results, improving the timeliness of information access for users.

[0075] In this embodiment, the development task is executed by calling the execution tool to obtain the execution result of the development task, so that the development task can be executed smoothly.

[0076] For multiple development task scenarios, when the development tasks include project creation tasks, device configuration tasks, and program writing tasks, step S30 includes: responding to the trigger operation of the control associated with the project creation task, determining the first execution tool corresponding to the project creation task in the integrated development environment; calling the first execution tool to execute the project creation task, and obtaining the execution result of the project creation task; displaying the execution result of the project creation task on the project screen, and displaying the response result of the execution result of the project creation task in the second area; responding to the trigger operation of the control associated with the device configuration task, determining the second execution tool corresponding to the device configuration task in the integrated development environment; calling the second execution tool to execute the device configuration task, and obtaining the execution result of the device configuration task; displaying the execution result of the device configuration task on the project screen, and displaying the response result of the execution result of the device configuration task in the second area; responding to the trigger operation of the control associated with the program writing task, determining the third execution tool corresponding to the program writing task in the integrated development environment; calling the third execution tool to execute the program writing task, and obtaining the execution result of the program writing task; displaying the execution result of the program writing task on the project screen, and displaying the response result of the execution result of the program writing task in the second area.

[0077] The aforementioned controls can be triggered automatically by the system. The entire task processing flow is automated after responding to the sending of the development requirement description information, making task processing more intelligent and efficient. Alternatively, the controls can be triggered by developers, who can activate the corresponding controls based on the actual situation to advance the task processing progress.

[0078] Furthermore, in one feasible implementation, it includes: Step S10: In the display interface, the dialogue screen of the intelligent assistant and the engineering screen of the target integrated development environment are displayed, wherein the dialogue screen includes a first area and a second area. Step S20: In response to the sending operation of the development requirement description information in the first area, the AI ​​big model is displayed in the second area to plan the development task based on the context information associated with the development requirement description information and the target project instance. Step S31: Respond to the trigger operation of the control associated with the development task and obtain the execution tool corresponding to the development task in the integrated development environment; Step S32: Invoke the execution tool to perform the development task; Step S331: If the execution tool call fails, the second area displays the first response result; or, the file acquisition tool is called to obtain the log file, the log file is parsed, and the first response result is obtained; the first response result is then displayed in the second area. Step S332: If the tool call is successful, the second response result will be displayed in the second area.

[0079] The execution result response includes a first response and a second response. The first response indicates execution failure, and may include the name of the tool that failed to be invoked and the reason for the failure. The second response indicates execution success, and may include messages such as "Execution successful: Project has been created" or "Execution successful: Main site creation completed."

[0080] In one feasible approach, when the intelligent assistant detects a failed execution tool call, it invokes a third-party file retrieval tool to obtain a log file. This log file stores all data involved in the execution of the development task. By parsing this log file, a first response result is obtained. The intelligent assistant can request an AI model to parse the log file to obtain the first response result. The AI ​​model can utilize relevant knowledge to parse and analyze the log file to obtain the first response result. This first response result is displayed in a second area, while the IDE maintains the current project view.

[0081] Alternatively, if the smart assistant detects a failed tool call, it can directly display the first response result in the second area, including the name of the failed tool and the reason for the failure.

[0082] In another feasible approach, upon detecting a successful execution tool call, the intelligent assistant displays messages such as "Execution successful: Project created" or "Execution successful: Main site creation complete" in the second area. At this point, the IDE then redirects to display the screen showing the execution status of the development task; for example, if the development task is to create a new project, the IDE will redirect to display the screen showing the creation of the new project.

[0083] In this embodiment, if the execution tool fails to call, a file retrieval tool can be used to obtain and parse the log file to determine the execution result, and this result is then displayed in the second area. Furthermore, it can display corresponding response results for different execution tool call scenarios, enabling developers to quickly locate and handle faults promptly, thus improving project development efficiency.

[0084] In other embodiments, the display interface shows a dialogue screen of the intelligent assistant and a project screen of the target integrated development environment (IDE). The dialogue screen includes a first area and a second area. Responding to the sending operation of development requirement description information input in the first area, context information associated with the target project instance is obtained; the development requirement description information and context information are integrated to construct a first prompt; the first prompt is sent to the AI ​​model, where the AI ​​model performs inference based on the first prompt to obtain the development task; the planned development task is displayed in the second area. Responding to the triggering operation of the control associated with the development task, the execution tool corresponding to the development task in the IDE is obtained; the execution tool is invoked to execute the development task; if the execution tool invocation fails, a file retrieval tool is invoked to obtain the log file; the log file is parsed to obtain a first response result; the first response result is displayed in the second area; if the execution tool invocation succeeds, a second response result is displayed in the second area. In this way, task planning, task execution, and result display are performed through a visual interface, providing convenience for task processing.

[0085] Based on the above embodiments of this application, in the second embodiment of this application, the content that is the same as or similar to the above embodiments can be referred to the above description, and will not be repeated hereafter. Furthermore, the task processing method based on large AI models also includes: Step S10: In the display interface, the dialogue screen of the intelligent assistant and the engineering screen of the target integrated development environment are displayed, wherein the dialogue screen includes a first area and a second area. Step S20: In response to the sending operation of the development requirement description information in the first area, the AI ​​big model is displayed in the second area to plan the development task based on the context information associated with the development requirement description information and the target project instance. Step S31: Respond to the trigger operation of the control associated with the development task and obtain the execution tool corresponding to the development task in the integrated development environment; Step S32: Invoke the execution tool to perform the development task; Step S110: If the execution tool call fails, integrate the response result, context information, development task, and execution result of the development task to obtain the second prompt message; The response result here refers to the response result when the execution tool call fails, including the name of the tool that failed to be called and the reason for the failure.

[0086] The second prompt information is a type of input content used to convey specific task-related information to the AI ​​model. It integrates response results, contextual information, development tasks, and the execution results of development tasks, aiming to provide the AI ​​model with comprehensive and accurate information so that it can reason based on this information and generate development tasks that meet the requirements.

[0087] The context information here can be the same as the context information used to construct the first prompt information, or it can be obtained by updating the context information used to construct the first prompt information.

[0088] In one feasible approach, the response result, context information, development task, and execution result of the development task can be concatenated according to a certain format and then formatted to make it a text format that meets the input requirements of large AI models.

[0089] In another feasible approach, information fusion can be achieved using natural language processing (NLP) techniques. This involves employing NLP techniques such as information extraction and semantic understanding to perform in-depth analysis of response results, contextual information, development tasks, and the execution results of those tasks. Key information is extracted, the semantic relationships between them are understood, and then, based on these analysis results, the information is reorganized in natural language to construct a more accurate and clear second prompt.

[0090] Step S120: The second prompt information is sent to the AI ​​big model, whereby the AI ​​big model performs inference based on the second prompt information and adjusts the development task. Step S130: Display the adjusted development task in the second area.

[0091] In step S140, in response to the triggering operation of the control associated with the adjusted development task, the execution result of the adjusted development task is displayed on the project screen, and the response result of the execution result of the adjusted development task is displayed in the second area.

[0092] In this embodiment, by integrating the response result, context information, development task, and execution result of the development task when the execution tool fails, a second prompt message is obtained. The AI ​​big model is then invoked to re-perform reasoning based on the second prompt message in order to adjust the development task, correct its deviation, and improve its accuracy.

[0093] In other embodiments, the display interface shows a dialogue screen of the intelligent assistant and a project screen of the target integrated development environment (IDE). The dialogue screen includes a first area and a second area. In response to the sending operation of development requirement description information input in the first area, context information associated with the target project instance is obtained. The development requirement description information and context information are integrated to construct a first prompt message. The first prompt message is sent to the AI ​​model, where the AI ​​model performs inference based on the first prompt message to obtain a development task. The planned development task is displayed in the second area. In response to the triggering operation of the control associated with the development task, the execution tool corresponding to the development task in the IDE is obtained. The execution tool is invoked to execute the development task. If the execution tool invocation fails, the response result, context information, development task, and execution result of the development task are integrated to obtain a second prompt message. The second prompt message is sent to the AI ​​model, where the AI ​​model performs inference based on the second prompt message to adjust the development task. The adjusted development task is displayed in the second area. In response to the triggering operation of the control associated with the adjusted development task, the execution result of the adjusted development task is displayed in the project screen, and the response result of the execution result of the adjusted development task is displayed in the second area. This method supports task planning correction and improves the accuracy of task execution results.

[0094] Based on the above embodiments of this application, in the third embodiment of this application, the content that is the same as or similar to the above embodiments can be referred to the above description, and will not be repeated hereafter. Based on this, after step S30, the task processing method based on the AI ​​large model further includes: Step S210: If there are unexecuted development tasks, correct the unexecuted development tasks based on the execution results of the executed development tasks and the context information associated with the target project instance. The context information here can be the context information mentioned above, or it can be obtained by updating the context information mentioned above. As the task is executed, this context information will be updated synchronously.

[0095] In one feasible approach, a third prompt is generated based on the execution results of the executed development tasks and the contextual information associated with the target project instance. This third prompt is then sent to the AI ​​big model, which uses the third prompt to infer and obtain a replanned development task. This replanned development task is then used to correct the unexecuted development tasks. In this way, the initially planned development tasks can be corrected based on the actual running conditions, making the corrected development tasks more in line with project requirements.

[0096] In step S220, in response to the triggering operation of the control associated with the corrected unexecuted development task, the execution result of the corrected unexecuted development task is displayed on the project screen, and the response result of the execution result of the corrected unexecuted development task is displayed in the second area.

[0097] In this embodiment, the unexecuted development tasks are corrected by using the execution results of the executed development tasks and the context information associated with the target project instance. In this way, the development tasks obtained from the initial planning can be corrected according to the actual operation, thereby correcting the development tasks and making the corrected development tasks more in line with the project requirements, thus improving the accuracy of the planned development tasks.

[0098] Based on the above embodiments of this application, in the fourth embodiment of this application, the content that is the same as or similar to the above embodiments can be referred to the above description, and will not be repeated hereafter. Based on this, after step S30, the task processing method based on the AI ​​large model further includes: Step S310: If all development tasks have been completed, invoke the email tool based on the development requirement description information.

[0099] Whether the email tool is invoked depends on whether the development requirements description includes a request to send emails. If the development requirements description includes a request to send emails, the email tool can be invoked after all development tasks have been completed.

[0100] In step S320, the email tool generates the target email based on the execution result and the response result of the execution result, and pushes the target email.

[0101] The target email is an email automatically created by an email tool for a specific purpose, based on the execution results and responses obtained after all development tasks have been completed. It typically contains key information about the development task execution, such as whether the task was successfully completed, errors or warnings encountered during execution, etc., aiming to accurately communicate the development task execution status to relevant personnel, such as project managers, development team members, and clients, so that they can understand the project progress in a timely manner, make decisions, or arrange subsequent work.

[0102] In one feasible approach, an email template can be pre-designed, containing placeholders for filling in the specific content of the execution and response results. The email tool reads the execution and response data of all development tasks, fills this data into the corresponding placeholders in the email template according to a predefined format, and generates a complete target email. Then, the email tool calls the email sending interface to push the target email to the specified recipient's inbox.

[0103] In another possible approach, the email tool calls an API provided by the email service provider, passing the execution and response results as parameters. The email service then generates the target email based on the received parameters, according to its internal rules and templates, and is responsible for pushing the email to the designated recipients.

[0104] In another feasible approach, once all development tasks are completed, an AI-powered large-scale model can be invoked to generate an email, which can then be pushed to designated recipients via email tools. The AI-powered large-scale model can generate emails based on task execution results, task response results, email subject lines, and recipient information.

[0105] In this embodiment, since a third-party email tool can be called to generate the target email, the automatically generated target email contains key information about the execution of the development task. Relevant personnel no longer need to spend time collecting and organizing information from multiple channels. They can simply check the email to fully understand the task execution status, thus improving communication efficiency.

[0106] Based on the above embodiments of this application, in the fifth embodiment of this application, the same or similar content as the above embodiments can be referred to the above description, and will not be repeated hereafter. On this basis, before step S10, the task processing method based on the AI ​​large model may further include: Step S410: In response to the scanning operation of integrated development environment instances, all scanned integrated development environment instances are displayed in the second area.

[0107] Step S420: In response to a trigger operation for any one of the integrated development environment instances among all integrated development environment instances, determine the target integrated development environment instance.

[0108] In one feasible approach, each integrated development environment (IDE) instance listens for external services through a preset port range. Simultaneously, the intelligent assistant continuously scans for IDE instances within this preset port range via a background task, records these instances, and displays them in the assistant's chat window, allowing the user to intuitively see which IDE instances are currently available. The user can select an IDE instance to interact with in the chat window. The intelligent assistant generates a request based on the selected IDE instance's listening parameters and sends it to the MCP server. Upon receiving the parameterized request, the MCP server locates the corresponding IDE instance based on the port number in the parameters and then invokes the services of that IDE instance through that port to perform subsequent operations.

[0109] The preset port range can be 9000~9020.

[0110] The listening parameters mentioned above can be parameters such as port number, window handle, and integrated development environment instance name.

[0111] In this embodiment, by scanning the integrated development environments (IDEs) in a listening state and displaying them in the second area, developers can quickly understand all currently running IDE instances or those requiring intelligent assistant task processing in a centralized interface. This enables one intelligent assistant to support the development needs of multiple IDE instances without needing to adapt a corresponding intelligent assistant for each IDE instance, thus reducing development costs and improving development efficiency.

[0112] It should be noted that, referring to Figure 2 and Figure 3 This application supports the development of multiple IDE terminals. A single IDE MCP server can schedule multiple IDE terminals, eliminating the need to deploy AI models or MCP modules separately for each IDE, thus reducing coupling and resource consumption. Multiple IDE terminals refer to dedicated IDE instances for PLC development, such as industrial IDEs that support device configuration, POU programming, and axis group management. Each terminal automatically starts and listens for services via a port, serving as the platform for operations such as "creating a new project, adding an etherCAT master station, and writing axis control programs." Its key features include built-in IDE project context information, support for Web API calls, and dynamic scanning and identification.

[0113] The smart assistant client contains the following components: Service API: This is a core module of the smart assistant client. It serves as the intermediate scheduling layer between the front-end UI, the MCP server, and the AI ​​big model. It is responsible for three major functions: session management, task planning, and tool execution. It connects with user requests from the front-end UI, forwards tool call instructions to the MCP server, and transmits task requirements and execution results to the AI ​​big model.

[0114] MCP servers are divided into IDE MCP servers and Other MCP servers, such as file services and email services. IDE MCP servers specifically handle PLC IDE-related operations, such as `read_pou` and `create_project`; Other MCP servers provide third-party service extensions. Based on the standardized MCP protocol interface, they support routed calls according to IDE listening parameters, which is crucial for multi-terminal adaptation.

[0115] MCP Client: Integrated into the smart assistant client, it is the entry point for initiating MCP protocol requests. It is responsible for encapsulating the service API instructions into the MCP standard format and attaching IDE listening parameters such as port and process ID, and passing them to the MCP server to ensure that the instructions conform to the MCP protocol specifications.

[0116] Front-end UI: This is the interactive interface of the intelligent assistant client. It is the direct interaction layer between the user and the system, providing operation entry points such as session windows, project management, configuration management, and device management. It displays a list of multiple IDE terminals for users to select the target IDE, presents the task steps and execution results planned by AI, and receives user requests.

[0117] For example, to help understand the implementation flow of the task processing method based on the AI ​​large model obtained by combining the above embodiments, refer to... Figures 2-4 Taking "a user requests to write a program with 6 axes" as an example, the complete workflow is divided into 5 main stages: (1) Discovery and list display of multiple IDE terminals.

[0118] The IDE terminal starts automatically. Multiple PLC IDE instances start and listen to services through ports 9000~9020 respectively. For example, IDE1 occupies port 9001 and IDE2 occupies port 9002. Each terminal automatically generates listening parameters including port number, process ID and IDE title. The underlying module of the smart assistant client periodically scans ports 9000-9020 to collect listening parameters and project context information of all active IDE terminals and compiles them into an IDE list. The service API pushes the IDE list to the front-end UI, allowing users to intuitively select the target project instance and avoid chaotic calls across multiple terminals.

[0119] (2) Initiate user needs and plan AI tasks.

[0120] Users send development requirement descriptions through the front-end UI, such as "write a program with 6 axes", and the front-end UI passes the requirements to the service API; The service API triggers an IDE context retrieval command, which is encapsulated into an MCP request by the MCP client and sent to the IDE MCP server. The IDE MCP server calls the Web API of the target project instance to obtain the IDE's project context information, such as the current project's hardware model, configured variables, and axis group parameters, and sends it back to the service API. The service API passes development requirement descriptions and project context information to the AI ​​model. Based on PLC industrial scenario constraints, such as the need for a 6-axis PLC to rely on an EtherCAT master station and to match a specific POU structure, the AI ​​model breaks down these constraints into executable task sequences, for example: Task 1: Create a new PLC project; Task 2: Add the etherCAT master station; Task 3: Create a 6-axis axis group; Task 4: Write ST language axis control program; The AI ​​big data model sends the planned development task sequence back to the service API, which then displays it to the user through the front-end UI.

[0121] (3) MCP protocol call and IDE task execution.

[0122] After the user confirms the execution, the service API processes the tasks one by one according to the development task sequence, first passing the "Task 1: Create a new project" instruction to the MCP client; The MCP client encapsulates the commands into the MCP standard format and attaches the target project instance's listening parameters, such as port 9001 and process ID=1234, to ensure that the MCP server can accurately locate the target project instance. The IDE MCP server receives the MCP request, parses the listening parameters, and calls the new project interface of the target project instance through the Web API. The IDE terminal executes the operation and returns a result indicating successful creation. The IDE MCP server sends the successful creation result back to the service API through the MCP client, and the service API then displays it to the user through the front-end UI.

[0123] (4) AI iteration planning and task closed loop.

[0124] The service API passes the result of the successful execution of Task 1 to the AI ​​model. The model, combined with the project context, determines that "the next step is to execute Task 2: add the etherCAT master station". Iterative execution of tasks: Repeat step (3) to execute tasks such as adding etherCAT master station, creating 6-axis axis group, and writing ST program in sequence. The results of each step are sent back to the AI ​​large model. Once the final step of writing the 6-axis program is successfully executed, the AI ​​big model verifies all task execution results and engineering context. If there are no syntax errors in the program and the axis group matches the hardware, it determines that all tasks have been completed and provides the final results to the user through the service API and front-end UI.

[0125] (5) Third-party service extension.

[0126] If a user needs to export the 6-axis program as a file, the front-end UI initiates an export file request, and the service API calls the MCP client. The MCP client encapsulates the file export instructions and passes them to another MCP server, such as the File service. The File service calls the file export interface of the target project instance through the Web API, generates a local file, and then sends the successful export result back to the front-end UI.

[0127] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the task processing method based on the AI ​​large model of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0128] Based on the same inventive concept, this application provides a task processing device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the task processing method based on the AI ​​large model in the above embodiments.

[0129] The task processing device provided in this application, employing the AI-based large model task processing method described in the above embodiments, can improve project development efficiency and shorten project development cycle. Compared with the prior art, the beneficial effects of the task processing device provided in this application are the same as those of the AI-based large model task processing method provided in the above embodiments, and other technical features of this task processing device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0130] Based on the same inventive concept, this application provides a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, which are used to execute the task processing method based on the AI ​​large model in the above embodiments.

[0131] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory (EPROM), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, radio frequency (RF), etc., or any suitable combination thereof.

[0132] The aforementioned computer-readable storage medium may be included in the task processing device or may exist independently and not assembled into the task processing device.

[0133] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by a task processing device, enable the task processing device to improve the development efficiency of the project and shorten the development cycle of the project.

[0134] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0135] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0136] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0137] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., computer programs) for executing the above-described task processing method based on a large AI model. This improves project development efficiency and shortens the project development cycle. Compared with related technologies, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the task processing method based on a large AI model provided in the above embodiments, and will not be repeated here.

[0138] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A task processing method based on a large AI model, characterized in that, The method includes: The display interface shows the dialogue screen of the intelligent assistant and the engineering screen of the target integrated development environment, wherein the dialogue screen includes a first area and a second area. In response to the sending operation of the development requirement description information in the first area, the AI ​​big model is displayed in the second area to plan the development task based on the context information associated with the development requirement description information and the target project instance; In response to the triggering operation of the control associated with the development task, the execution result of the development task is displayed on the project screen, and the response result of the execution result is displayed in the second area.

2. The task processing method based on a large AI model as described in claim 1, characterized in that, In response to the sending operation of the development requirement description information input in the first area, the AI ​​big model is displayed in the second area based on the context information associated with the development requirement description information and the target project instance, and the planned development tasks include: In response to the sending operation of the development requirement description information input in the first area, obtain the context information associated with the target project instance; Integrate the development requirement description information and the context information to construct the first prompt information; The first prompt information is sent to the AI ​​big model, wherein the AI ​​big model performs inference based on the first prompt information to obtain the development task; The development tasks obtained from the planning are displayed in the second area.

3. The task processing method based on a large AI model as described in claim 1, characterized in that, The response to the triggering operation of the control associated with the development task, displaying the execution result of the development task on the project screen, and displaying the response result of the execution result in the second area includes: In response to the triggering operation of the control associated with the development task, obtain the execution tool corresponding to the development task in the integrated development environment; The execution tool is invoked to perform the development task; The execution result of the development task is displayed on the engineering screen, and the response result of the execution result is displayed in the second area.

4. The task processing method based on a large AI model as described in claim 3, characterized in that, The response result of displaying the execution result in the second area includes: If the execution tool call fails, the second area displays the first response result; or, the file acquisition tool is called to obtain the log file, the log file is parsed to obtain the first response result, and the first response result is displayed in the second area. If the execution tool is successfully invoked, the second response result will be displayed in the second area; The response result includes either the first response result or the second response result.

5. The task processing method based on a large AI model as described in claim 3, characterized in that, After the execution tool is invoked to perform the development task, the process further includes: If the execution tool call fails, the response result, the context information, the development task, and the execution result of the development task are integrated to obtain a second prompt message; The second prompt message is sent to the AI ​​big model, wherein the AI ​​big model performs inference based on the second prompt message and adjusts the development task accordingly; The adjusted development tasks are displayed in the second area; In response to the triggering operation of the control associated with the adjusted development task, the execution result of the adjusted development task is displayed on the project screen, and the response result of the execution result of the adjusted development task is displayed in the second area.

6. The task processing method based on a large AI model as described in claim 1, characterized in that, The method of responding to the triggering operation of the control associated with the development task, after displaying the execution result of the development task on the project screen and displaying the response result of the execution result in the second area, further includes: If there are unexecuted development tasks, the unexecuted development tasks are corrected based on the execution results of the executed development tasks and the context information associated with the target project instance. In response to the triggering operation of the control associated with the corrected unexecuted development task, the execution result of the corrected unexecuted development task is displayed on the project screen, and the response result of the execution result of the corrected unexecuted development task is displayed in the second area.

7. The task processing method based on a large AI model as described in claim 1, characterized in that, The method of responding to the triggering operation of the control associated with the development task, after displaying the execution result of the development task on the project screen and displaying the response result of the execution result in the second area, further includes: If all development tasks are completed, the email tool will be invoked based on the development requirement description information. Based on the execution result and the response result of the execution result, a target email is generated and the target email is pushed.

8. The task processing method based on a large AI model as described in any one of claims 1 to 7, characterized in that, The task processing method based on the AI ​​large model also includes: In response to the scanning operation of integrated development environment instances, all scanned integrated development environment instances are displayed in the second area; In response to a triggered operation targeting any one of the said integrated development environment instances, the target integrated development environment instance is determined.

9. A task processing device, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the task processing method based on a large AI model as described in any one of claims 1 to 8.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the task processing method based on a large AI model as described in any one of claims 1 to 8.

Images