A data processing method, apparatus and device, and a storage medium

Abstract

Embodiments of the present application provide a data processing method, device and equipment, and a storage medium. The method comprises: displaying a first conversation message input by a conversation object in a conversation page corresponding to an intelligent agent; displaying a second conversation message for replying to the first conversation message in the conversation page; the second conversation message is generated by the intelligent agent calling a first applet to execute a first business task, and the first business task is a business task indicated by the first conversation message. The calling efficiency of the applet can be improved by the present application.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a data processing method, apparatus, device, and storage medium. Background Technology

[0002] With the rapid development of the mobile internet, mini-programs have quickly become popular as a lightweight form of entertainment that can be used instantly. In practice, it has been found that because users are not familiar with the functions of various mini-programs, they need to try multiple times to find the right one, resulting in relatively low efficiency in calling mini-programs. Summary of the Invention

[0003] This application provides a data processing method, apparatus, device, and storage medium that can improve the efficiency of calling mini-programs.

[0004] One embodiment of this application provides a data processing method, including: The first conversation message input by the conversation object is displayed on the conversation page corresponding to the intelligent agent; The conversation page displays a second conversation message used to reply to the first conversation message; the second conversation message is generated by the agent calling the first mini-program to execute the first business task, and the first business task is the business task indicated by the first conversation message.

[0005] One embodiment of this application provides a data processing method, including: The first conversation message input by the conversation object is displayed on the conversation page corresponding to the intelligent agent; Retrieve the program description information of candidate mini programs in the mini program list; The intelligent agent determines a first mini-program from the mini-program list to perform the first business task based on the program description information; the first business task is the business task indicated by the first session message. Through the aforementioned intelligence, the first mini-program is invoked to execute the first business task, thereby generating a second session message for replying to the first session message; The second session message is displayed on the session page.

[0006] One embodiment of this application provides a data processing apparatus, including: The first display module is used to display the first conversation message input by the conversation object in the conversation page corresponding to the intelligent agent; The second display module is used to display a second session message in the session page as a reply to the first session message; the second session message is generated by the agent calling the first mini-program to execute the first business task, and the first business task is the business task indicated by the first session message.

[0007] One embodiment of this application provides a data processing apparatus, including: The first display module is used to display the first conversation message input by the conversation object in the conversation page corresponding to the intelligent agent; The acquisition module is used to obtain the program description information of candidate mini programs in the mini program list; The determining module is configured to, through the intelligent agent, determine, from the list of mini-programs a first mini-program for performing a first business task based on the program description information; the first business task is the business task indicated by the first session message. The calling module is used to invoke the first mini-program to execute the first business task through the intelligence, so as to generate a second session message for replying to the first session message; A second display module is used to display the second session message on the session page. One embodiment of this application provides a computer device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0008] One embodiment of this application provides a computer storage medium storing a computer program, which, when executed by a processor, performs the steps of the above-described method.

[0009] One aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0010] In this application, the first mini-program is invoked by the intelligent agent to execute the first business task indicated by the first session message, so as to generate a second session message for replying to the first session message. This eliminates the need for the user to manually select the mini-program to run after multiple attempts, improves the invocation efficiency of the mini-program, and provides users with more efficient and convenient business tasks. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the structure of a data processing system provided in this application; Figure 2 This is a schematic diagram of the structure of a data processing system provided in this application; Figure 3 This is a flowchart illustrating a data processing method provided in this application; Figure 4 This is a schematic diagram of an interface provided in this application for calling a single mini-program to return a session message; Figure 5 This is a schematic diagram of an interface provided in this application for calling multiple mini-programs to return session messages; Figure 6 This is a schematic diagram of an interface for displaying authorization request information provided in this application; Figure 7 This is a schematic diagram of an interface provided in this application, showing how a user selects a first mini-program; Figure 8 This is a schematic diagram of an interface for displaying change messages provided in this application; Figure 9 This is a schematic diagram of the conversation page of a mini-program assistant provided in this application; Figure 10 This is a flowchart illustrating yet another data processing method provided in this application; Figure 11 This is a schematic diagram of the structure of a data processing device provided in this application; Figure 12 This is a schematic diagram of the structure of a data processing device provided in this application; Figure 13 This is a schematic diagram of the structure of a computer device provided in this application. Detailed Implementation

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0013] First, let me explain the terms used in this application: (1) Intelligent agent: an autonomous decision-making and execution entity driven by a large language model (LLM), capable of understanding user intent, selecting and calling tools, and returning execution results. The intelligent agent in this application can be used to call and execute mini-programs to provide business tasks for users.

[0014] The business tasks may include weather forecasting, ride-hailing, ticket booking, information retrieval, and so on.

[0015] (2) Mini Program: A lightweight, closed, efficient, and secure program based on Web technology that runs in a specific host environment (host application). The host application can refer to the program that provides the runtime environment for the mini program. The host application needs to download and install the package to run, while the mini program can run without downloading and installing the package.

[0016] Mini programs can include social mini programs, game mini programs, shopping mini programs, payment mini programs, ride-hailing mini programs, etc.

[0017] (3) Atomic interface: It is the smallest execution unit in the mini-program, encapsulates a single business function, has standardized input parameters and output structure, and runs in a sandbox environment. The sandbox environment may refer to the first-level sub-security operating environment in this application. For an explanation of the first-level sub-security operating environment, please refer to the description below.

[0018] (4) Atomic Component: A visual display unit for atomic interfaces, which renders the structured data returned by the atomic interface into a graphical user interface (GUI) card, a traditional visual operation interface.

[0019] To facilitate a clearer understanding of this application, we will first introduce... Figure 1 , Figure 1 This is a schematic diagram of the structure of a data processing system provided in this application. Figure 1 As shown in the diagram, the data processing system may include a server and a terminal cluster, which may include multiple terminals. Figure 1 As shown, taking a terminal cluster consisting of three terminals as an example, the terminal cluster can specifically include a first terminal, a second terminal, and a third terminal. It can be understood that the first terminal, the second terminal, and the third terminal can all connect to the server via the network so that each terminal can interact with the server through the network connection.

[0020] Understandably, a server can be a single physical server, a server cluster or distributed system consisting of at least two physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud knowledge bases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. A terminal can specifically refer to in-vehicle terminals, smartphones, tablets, laptops, desktop computers, smart speakers, speakers with screens, smart TVs, smartwatches, etc., but is not limited to these. Various terminals and servers can be directly or indirectly connected via wired or wireless communication. Furthermore, the number of terminals and servers can be one or at least two; this application does not impose any restrictions.

[0021] The terminal may include a mini-program assistant, which can be a standalone application, a web application, or a mini-program within a host application. This mini-program assistant can serve as a front-end client for the intelligent agent. It can acquire user-inputted conversational messages, input these messages into the intelligent agent, and then the intelligent agent invokes the corresponding mini-program to generate a response message.

[0022] The terminal also has one or more programs installed, which can be called host applications. Each host application can be used to provide a runtime environment for the mini-program, and the mini-program runs in the runtime environment to perform business tasks for the user.

[0023] A server can refer to a device that provides backend services for a mini-program or a host application. For example, it may verify a user's access request and allow the user to access the mini-program if the user has the necessary access permissions.

[0024] like Figure 2 The diagram shown is an architectural schematic of a data processing system provided in this application. The data processing system includes a client and a backend. The client can refer to a host application, which provides a user interaction entry point and a mini-program runtime environment. The user interaction entry point includes user input of natural language, account connection, mini-program link, and text + atomic components (i.e., atomic rendering components).

[0025] The mini-program runtime environment can realize functions such as login, payment, and geolocation, and can be used to run the atomic services of the mini-program.

[0026] The backend can refer to the backend of the host application, which includes the mini-program assistant, used for intent understanding, task planning, task scheduling, session management, memory system, etc.

[0027] The backend also includes a recall system, a knowledge base, and skills and application descriptions for N mini-programs (i.e., candidate mini-programs).

[0028] Here, "skills" can refer to the functional data package that enables a mini-program to complete specific business tasks. This functional data package includes usage instructions (SKILL.md), function declarations, atomic interfaces, and atomic components.

[0029] In one embodiment, a mini-program can perform multiple business tasks, and the functional data package is a scenario-oriented complete capability encapsulation, which allows the various business tasks provided by the mini-program to be split into multiple functional data packages. For example, a ticket booking mini-program can perform the business of booking air tickets, train tickets, and bus tickets, and can generate corresponding functional data packages for booking air tickets, train tickets, and bus tickets respectively.

[0030] Each feature package includes: a user manual, a callable feature declaration (mcp.json) displayed to the large language model, atomic interfaces, and atomic components.

[0031] The user manual includes: a functional specification document for the large language model in the agent to understand, including function name and description, usage scenario description, input and output format, calling examples, etc., so that the agent can accurately understand the applicable scenarios and calling methods of the function during the execution phase.

[0032] Among them, the callable function declarations provided to the model (mcp.json) are: the callable capabilities are taken over by atomic interfaces, and the large language model can orchestrate and schedule the atomic interfaces declared here when using the function data package.

[0033] Among them, the feature declaration is a declarative definition of atomic interfaces through mcp.json, and the feature declaration aligns with the three major capability lists of the standard Model Context Protocol (MCP).

[0034] This functional declaration includes interface declaration fields (apis), component declaration fields (components), and prompt declaration fields (prompts). The interface declaration fields (apis) correspond to the tools / list in the standard model context protocol and are used to declare atomic interfaces (tools), including the name of the atomic interface, its input and output formats, etc. The component declaration fields correspond to the resources / list in the standard model context protocol and are used to declare atomic components. The interface declaration fields are used to declare prompt templates.

[0035] It should be noted that developers can integrate the feature through the standard process of "declaring the feature, writing SKILL.md, declaring the MCP, implementing the atomic interface, and implementing the atomic component", without having to refactor the existing mini-program code.

[0036] The recall system is used for service recall, page recall, and account recall, while the knowledge base is used for account semantic understanding and page semantic understanding.

[0037] The data processing system refers to a three-stage pipeline architecture. Phase 1: Recall Phase; Input: User request (first session message), object attribute information, historical preferences; Processing: Recall the most relevant mini-programs and their function data packages from the mini-program list and corresponding function library; Output: Specific function declaration + corresponding MCP tool declaration (tools list).

[0038] Phase Two: Execution Phase; Input: User request + Function description + Recall tool declaration + Context information; Core loop: Organize the current context; The large language model in the agent makes decisions to generate the calling instruction; The target atomic interface executes in the sandbox; Determine whether to continue calling the tool or return the result; Output: When it is not necessary to continue calling the tool, the result is executed by the target atomic interface.

[0039] Phase 3: Summary Phase; Input: Execution results from the execution phase; Processing: Transforming the execution results into user-friendly natural language responses or GUI card rendering (atomic rendering component); Output: User-oriented natural language text or GUI cards.

[0040] Please see Figure 3 , Figure 3 This is a flowchart illustrating a data processing method provided in this application. Figure 3 As shown, this method can be derived from... Figure 1 The method, executed via a terminal in the system, may include the following steps: S101. In the conversation page corresponding to the agent, the first conversation message input by the conversation object is displayed.

[0041] In this application, the terminal can display the first conversation message input by the conversation object on the conversation page corresponding to the smart agent.

[0042] The conversation page can be a page belonging to the mini-program assistant, which may be presented as a standalone application, a web application, or a mini-program within a host application.

[0043] The first session message can be used to instruct the agent on the first business task to be performed. The first business task may include, but is not limited to, weather query task, ride-hailing task, ticket booking task, information query task, etc.

[0044] In one embodiment, the terminal can display a floating window about the mini-program assistant entry point in the host application. In response to a trigger operation on the floating window, the mini-program assistant is launched, and its session page is displayed. This session page is identified as the session page corresponding to the intelligent agent. In response to an input operation on the session page, the entered first session message is displayed on the session page.

[0045] In one embodiment, the terminal can display a swipe operation on the host application's session page, display a mini-program assistant entry point, and, in response to a trigger operation on the mini-program assistant entry point, display the mini-program assistant's session page, identifying the mini-program assistant's session page as the session page corresponding to the intelligent agent. The swipe operation can refer to swipes from left to right, right to left, top to bottom, bottom to top, etc.

[0046] The first conversation message can be input by the conversation object via voice, manual input, etc., or it can refer to the message selected by the conversation object from among the candidate conversation messages.

[0047] Candidate conversation messages are conversation messages that are input into the agent at a frequency greater than a frequency threshold.

[0048] In one embodiment, the terminal can respond to a trigger operation on a smart mini-program associated with a smart agent by displaying the conversation page corresponding to the smart agent. That is, upon detecting a launch operation of the smart mini-program, the terminal can launch the smart mini-program and display its conversation page, which is the same as the conversation page corresponding to the smart agent. By calling the smart agent through the smart mini-program, and then using the smart agent to call various mini-programs, it is beneficial to provide users with efficient and convenient business tasks.

[0049] Here, the "mini-program" can refer to a mini-program assistant.

[0050] S102. On the aforementioned conversation page, a second conversation message is displayed to reply to the aforementioned first conversation message; the aforementioned second conversation message is generated by calling the first mini-program to execute the first business task, and the aforementioned first business task is the business task indicated by the aforementioned first conversation message.

[0051] In this application, after obtaining the first conversation message, a first mini-program can be invoked and executed by an intelligent agent to generate a second conversation message to reply to the first conversation message, and the second conversation message can be displayed on the conversation page.

[0052] The first mini-program is used to execute the first business task indicated by the first session message; the second session information is the execution result of the first business task.

[0053] In this context, when the first session message is used to instruct the agent to provide a weather query task, the first mini-program is the weather mini-program, and the second session message is the weather information queried in the weather mini-program.

[0054] In this context, when the first session message is used to instruct the agent to provide a ride-hailing task, the first mini-program is a ride-hailing mini-program, and the second session message is the information of the vehicle to be hailed in the ride-hailing mini-program.

[0055] In this context, when the first session message is used to instruct the agent to provide a ticket booking task, the first mini-program is the ticket booking mini-program; the second session message is the ticket information booked in the ticket booking mini-program.

[0056] In this context, when the first session message is used to instruct the agent to provide a query information task, the first mini-program is the information mini-program, and the second session message is the information queried in the information mini-program.

[0057] For example, such as Figure 4 As shown, the terminal can display a conversation page 41a between the conversation object and the smart object. This conversation page 41a displays the first conversation message 42a entered by the conversation object, which reads: "What's the weather like at location A today?" The smart agent determines that the weather app is the first app. The smart agent can call this weather app to check the weather for location A today. The terminal can display first execution progress information 43a, call prompt information 44a, second execution progress information 45a, and second conversation message 46a on the conversation page. The first execution progress information 43a reads: "Okay, checking the weather for location A for you." The call prompt information 44a reads: "The weather app is checking the weather for you." The second execution progress information 45a reads: "The weather for location A has been checked for you. Please check the details below." The second conversation message reads: "The weather at location A is sunny, 28 degrees Celsius."

[0058] Among them, the intelligent agent object can be the object controlled by the intelligent agent, that is, the conversation content output by the intelligent object is determined by the intelligent agent.

[0059] In one embodiment, the terminal may display a call prompt message about the first mini-program on the session page; the call prompt message is used to reflect that the agent is calling the first mini-program; this is beneficial for understanding the mini-program called by the agent in real time.

[0060] For example, if a ride-hailing mini-program is currently being used, the notification message could be: "The ride-hailing mini-program is calling a ride."

[0061] In one embodiment, the terminal can display execution progress information about the first mini-program on the session page; the execution progress information is used to reflect the progress of the first mini-program in executing the first business task, such as whether the first business task has been completed; this is beneficial for users to understand the execution progress of the first mini-program in real time.

[0062] For example, if a ride-hailing mini-program is currently being used, the progress information will be: "Calling a ride, 83 people nearby have called for a ride, please wait." Or, if a ticket booking mini-program is currently being used, the progress information will be: "Flight information to destination A has been found."

[0063] In one embodiment, the terminal can display a third conversation message input by the conversation object in response to the second conversation message on the conversation page; and display a fourth conversation message for replying to the third conversation message on the conversation page; that is, the intelligent agent can call multiple mini-programs to provide business tasks for the user.

[0064] The fourth session message is generated by the agent calling the second mini-program to execute the second business task according to the context information. The second business task is the business task indicated by the third session message. The context information includes at least one of the first session message, the second session message, or the third session message.

[0065] The second mini-program can be the first mini-program, meaning the agent can call the same mini-program to conduct multi-turn dialogues with the conversation object to perform business tasks for the user. Alternatively, the second mini-program can be different from the first mini-program, meaning the agent can call different mini-programs to perform business tasks for the user. For example, Figure 5As shown, the terminal can display a conversation page 51a between the smart object and the conversation object. This conversation page 51a includes a first conversation message 52a input by the conversation object: "Buy me some cold medicine." The smart agent can identify a first mini-program associated with this first conversation message 52a, which is a medicine mini-program that can be used to purchase medicine. The smart agent can call the medicine mini-program to search for cold medicine and generate a second conversation message 53a. This second conversation message 53a indicates that a list of cold medicines has been found in the medicine mini-program, including medicine 1, medicine 2, and medicine 3, as well as the price of each medicine. The conversation page 51a also includes a third conversation message 54a input by the conversation object: "Buy medicine 2." Based on the third conversation message 54a, the smart agent outputs a fourth conversation message 55a: "The medicine mini-program supports payment method 1 and payment method 2. Please select a payment method." The session page 51a also includes a fifth session message 56a input by the session object and a sixth session message 57a output by the intelligent agent. Session message 56a is: Payment method 2, and the sixth session message 57a is the payment order for drug 2 generated by the drug mini-program.

[0066] In one embodiment, the terminal can display a message card corresponding to the first mini-program on the session page; the message card displays a second session message for replying to the first session message, and the interaction with the user can be achieved through the message card.

[0067] The message card can be rendered by the native rendering component corresponding to the first mini-program.

[0068] In one embodiment, displaying a second session message in the session page as a reply to the first session message includes: the terminal can display authorization request information in the session page; the authorization request information is used to indicate whether to authorize the first mini-program to obtain the object attribute information of the session object; upon detecting a confirmation operation for the authorization request information, indicating that the session object agrees to use the object attribute information, in response to the confirmation operation, displaying a second session message in the session page as a reply to the first session message; the second session message is generated by the agent calling the first mini-program to execute the first business task based on the object attribute information. By returning the second session message only after authorization, based on the object attribute information and the first session message, information security is improved.

[0069] The second session message is generated by the agent calling the first mini-program to execute the first business task based on the object attribute information and the first session message.

[0070] The object's attribute information includes the session object's home address, current location, phone number, facial information, etc.

[0071] For example, such as Figure 6 As shown, the terminal can display a session page 61a, which includes a first session message 62a input by the session object. The first session message 62a is: "Help me hail a taxi to location B." The intelligent agent can determine a first mini-program for executing the first business task indicated by the first session message. This first mini-program is a taxi-hailing mini-program. When the taxi-hailing mini-program needs to obtain the location information of the session object during execution, the terminal can display authorization request information 63a on the session page 61a. This authorization request information 63a includes the taxi-hailing mini-program requesting to obtain your location information, an allow control, and a deny control. When a trigger operation for the deny control is detected, the taxi-hailing mini-program can be paused. When a trigger operation for the allow control is detected, an authorization prompt message is displayed on the session page 61a. This authorization prompt message instructs the session object to authorize the acquisition of its location information; specifically, the authorization prompt message is: "Allow the taxi-hailing mini-program to obtain your location information." The terminal can continue to invoke the taxi-hailing application through the intelligent agent, providing taxi services to the session object based on its location information and the first session message.

[0072] In one embodiment, the second session message may be displayed in a message card that allows interaction. The terminal may respond to a trigger operation on the second session message by displaying the business page of the first mini-program; the business page displays business content associated with the second session message; and the second session message enables interaction with the first mini-program.

[0073] For example, if an interaction with a message card generated by a drug mini-program is detected, the drug mini-program can be launched, its business page can be displayed, and the drug order can be shown on the business page.

[0074] In one embodiment, the terminal may display an image on the session page when the invocation time of the first mini-program exceeds a time threshold, or when the remaining memory resources are less than a first resource threshold; the image is obtained by taking a screenshot of a message card including the second session message. When the Context is recycled, the rendered message card is displayed in a downgraded manner through a screenshot (preserving a visual snapshot), and a cold start recovery is triggered when the user interacts again (recreating the Context and restoring its state), achieving memory optimization that is "unnoticeable to the user".

[0075] Context can refer to the runtime environment in which the first mini-program runs.

[0076] In one embodiment, when the remaining memory resources in the terminal are greater than or equal to a second resource threshold, the terminal can continue to display the session page and not reclaim the content in the session page.

[0077] In one embodiment, when the remaining memory resources in the terminal are greater than a first resource threshold and less than a second resource threshold, the terminal can delete information about sessions that are not visible on the page.

[0078] In one embodiment, in response to an interaction with the session page, the terminal restores the display of the message card including the second session message on the session page; that is, when the user interacts again, the Context is recreated and the display state of the message card for the second session message is restored.

[0079] In one embodiment, the terminal may display at least one candidate mini-program associated with the first session message on the session page; in response to a selection operation for any candidate mini-program among the at least one candidate mini-program, the selected mini-program is determined as the first mini-program that provides business tasks for the first session message. By having the user automatically select the first mini-program that provides business tasks for the first session message, it is beneficial to better provide business tasks to the user.

[0080] For example, such as Figure 7 As shown, the terminal can display a conversation page 71a, which includes a first conversation message 72a input by the conversation object. The first conversation message 72a is: "Help me hail a taxi to location B." The agent can obtain multiple candidate mini-programs associated with this first conversation message; that is, the terminal can display a candidate mini-program list 73a on the conversation page 71a. This candidate mini-program list 73a includes taxi-hailing mini-program 1, taxi-hailing mini-program 2, and taxi-hailing mini-program 3. In response to the selection operation for taxi-hailing mini-program 1, taxi-hailing mini-program 1 can be used as the first mini-program providing the business task in the first conversation message. The agent can call taxi-hailing mini-program 1 to provide taxi-hailing services to the user. The terminal can display execution progress information 74a about taxi-hailing mini-program 1 on the conversation page 71a. This execution progress information 74a is: "Taxi-hailing mini-program 1 has obtained your location and is matching pick-up points."

[0081] In one embodiment, during the process of the terminal responding to the first session message by calling a third mini-program, in response to a message input operation on the session page, the terminal displays a change message entered by the session object on the session page; the change message is used to indicate a change of mini-program. The first mini-program for performing the first business task is then re-determined from the mini-program list.

[0082] For example, such as Figure 8 As shown, the terminal can display a conversation page 81a, which includes a first conversation message 82a input by the conversation object. The first conversation message 82a is: "Help me call a taxi to location B." The agent can obtain a third mini-program associated with the first conversation message, which is a taxi-hailing mini-program 2. The terminal can display a call prompt message 83a and an execution progress message 84a about the taxi-hailing mini-program 2 on the conversation page 82a. The call prompt message 83a is: "Taxi-hailing mini-program 2 is calling a taxi." The execution progress message 84a is: "Calling a taxi, 103 people nearby have called a taxi, please wait." The conversation page 82a also includes a change message 85a input by the conversation object, which is: "It's taking too long, help me call a taxi in taxi-hailing mini-program 1 as well." Based on the update message 85a, the intelligent agent can call the ride-hailing mini-program 1 to provide ride-hailing services to the user, and display the call prompt message 86a of the ride-hailing mini-program 1 on the conversation page 81a. The call prompt message 86a is: Ride-hailing mini-program 1 is calling a car.

[0083] It should be noted that developers only need to follow three steps to integrate: Declare the function: Declare the skill (including MCP function declaration and atomic interface list) in app.json, and receive the business logic and call examples written by the developer for the usage instructions; Implement the atomic interface: Receive the execution function with standard input and output written by the developer; Implement the atomic component: Receive the atomic rendering component for visualization written by the developer.

[0084] In one embodiment, the terminal can display a call control for the mini-program assistant within the first mini-program; when a trigger operation is detected on the call control, the mini-program assistant's session page is displayed. The aforementioned display of a first session message input by the session object in the session page corresponding to the intelligent agent includes: in response to an input operation on the session page, displaying the first session message input by the session object in the mini-program assistant's session page. By displaying the mini-program assistant's call control within the mini-program, the first mini-program can be directly invoked, improving the ease of invoking the first mini-program.

[0085] For example, such as Figure 9 As shown, the terminal can display the business page 91a of the first mini-program. This business page 91a includes a call control 92a. When a trigger operation is detected for the call control 92a, the mini-program assistant's conversation page can be displayed. This conversation page displays the first conversation message 94a, which is input by the conversation object, i.e., "Help me hail a taxi to location B." The agent can invoke the first mini-program to execute the taxi-hailing task, thereby generating a second conversation message to reply to the first conversation message.

[0086] In this application, the first mini-program is invoked by the intelligent agent to execute the first business task indicated by the first session message, so as to generate a second session message for replying to the first session message. This eliminates the need for the user to manually select the mini-program to run after multiple attempts, improves the invocation efficiency of the mini-program, and provides users with more efficient and convenient business tasks.

[0087] Please see Figure 10 , Figure 10 This is a flowchart illustrating yet another data processing method provided in this application. For example... Figure 10 As shown, this method can be derived from... Figure 1 The method is executed by a terminal or server in a computer, and the device used to execute this application may be called a computer device. The method may include the following steps: S201. On the conversation page corresponding to the agent, the first conversation message input by the conversation object is displayed.

[0088] S202. Obtain the program description information of the candidate mini programs in the mini program list.

[0089] This list of mini-programs includes those registered in the smart agent, which can be referred to as candidate mini-programs.

[0090] S203. The intelligent agent determines a first mini-program from the mini-program list to perform the first business task based on the program description information; the first business task is the business task indicated by the first session message.

[0091] In this application, a computer device can use the intelligent agent to determine a first mini-program for instructing a first business task from the mini-program list based on the program description information.

[0092] The program description information may include functional descriptions of the candidate mini-program (functional declaration), name, version, etc., and may also be referred to as application description.

[0093] In one embodiment, an agent invokes a large language model to perform intent recognition on a first session message to obtain a first business task indicated by the first session message. Based on the first business task and the functional description data packet of candidate mini-programs, a first mini-program for executing the first business task is determined.

[0094] In one embodiment, determining the first mini-program that provides the business task for the first session message from the mini-program list based on the program description information, through the intelligent agent, includes: the computer device obtaining program description information of candidate mini-programs from the functional description data package of candidate mini-programs; selecting candidate mini-programs associated with the first session message from the mini-program list through the intelligent agent based on the program description information; when there are multiple selected candidate mini-programs, obtaining the historical call count of the session object for each selected candidate mini-program; and determining the first mini-program for performing the first business task from the selected candidate mini-programs based on the historical call count. This approach, combined with the user's usage habits regarding mini-programs, helps improve the accuracy of obtaining the first mini-program and facilitates obtaining a first mini-program that matches the user's usage habits, thereby improving the user experience.

[0095] S204. Through the intelligence, the first mini-program is invoked to execute the first business task to generate a second session message for replying to the first session message.

[0096] In this application, a computer device can invoke the first applet through an intelligent agent to execute a first business task in order to generate a second session message for replying to the first session message.

[0097] In one embodiment, the step of invoking the first mini-program to execute the first business task through the intelligence to generate a second session message for replying to the first session message includes: obtaining functional description information (i.e., functional declarations) of candidate atomic interfaces of the first mini-program from its functional data package; determining a target atomic interface for executing the first business task from the candidate atomic interfaces of the first mini-program based on the functional description information. The intelligence agent then invokes the target atomic interface to execute the first business task according to the usage instructions of the first mini-program's functional data package to generate a second session message for replying to the first session message. Selecting a suitable target atomic interface through the functional description information of each candidate atomic interface helps improve the accuracy of replies to the first session message.

[0098] Among them, the candidate atomic interface can refer to the encapsulation of various functional components of the first mini-program by the developer of the first mini-program. The candidate atomic interface is the atomic interface that the large language model in the intelligent agent can call in the functional data package of the first mini-program.

[0099] The functional description information of the candidate atomic interfaces reflects the functions of each candidate atomic interface.

[0100] In one embodiment, the step of invoking the first mini-program to execute the first business task through the intelligent agent to generate a second session message for replying to the first session message includes: in the main secure operating environment, the intelligent agent generates a call instruction for the target atomic interface according to the usage instructions of the first mini-program's function data package (i.e., the calling method of the target atomic interface); in the first-level sub-secure operating environment, the target atomic interface is invoked to execute the first business task according to the call instruction to generate a second session message for replying to the first session message. By invoking the target atomic interface in the first-level sub-secure operating environment, the execution of instructions beyond the scope of permissions by the target atomic interface is avoided, thus improving the security of mini-program invocation.

[0101] The functional data packages of each mini-program are stored in the main secure runtime environment, and the atomic interfaces and atomic rendering components of each candidate mini-program are registered in the main secure runtime environment.

[0102] In one embodiment, the agent can invoke a large language model, generate a call instruction according to the usage instructions of the first mini-program's functional data package (i.e., the calling method of the target atomic interface) and the first business task, and send the call instruction to the host application. The host application then sends the call instruction to the main secure runtime environment. A thread in the host secure runtime environment, upon determining the validity of the call instruction, forwards it to the first-level sub-secure runtime environment. In the first-level sub-secure runtime environment, the target atomic interface is invoked according to the call instruction to execute the first business task, thereby generating a second session message to reply to the first session message.

[0103] The legitimacy of the call instruction can be one of the following: the mini-program used to execute the call instruction is a mini-program registered in the main security environment, or the mini-program's functional data package matches the first business task.

[0104] Specifically, the target atomic interface can execute instructions within the first-level sub-safe operating environment and run within that environment, thus preventing processes outside the first-level sub-safe operating environment from modifying the target atomic interface and improving its accuracy.

[0105] In one embodiment, the step of executing the first business task by calling the target atomic interface through the intelligent agent to generate a second session message for replying to the first session message includes: the terminal can call the target atomic interface through the intelligent agent to execute the first business task and obtain the execution result. Through the intelligent agent, in a second-level sub-security operating environment, the atomic rendering component of the first mini-program is called; the atomic rendering component renders the execution result to obtain the second session message for replying to the first session message, that is, the execution result is rendered into a message card carrying the second session message. By calling the atomic rendering component in the second-level sub-security operating environment, it is isolated from the main logic of the intelligent agent, thereby improving the operational security of the intelligent agent.

[0106] In one embodiment, after the target atomic interface obtains the execution result, the execution result can be sent to the host application. The host application then sends the execution result to a large language model, which converts the execution result into natural language, resulting in a converted execution result. The intelligent agent then sends the converted execution result back to the host application, which in turn sends it to the atomic rendering component corresponding to the first mini-program. The atomic rendering component renders the converted execution result to obtain a second session message used to reply to the first session message. This interaction between the first mini-program and the intelligent agent is achieved through the host application, reusing the host application's existing secure channels and improving the security of the interaction between the mini-program and the intelligent agent.

[0107] The atomic rendering component can execute instructions within the second-level sub-safe runtime environment and run within that environment. This prevents processes outside the second-level sub-safe runtime environment from modifying the atomic rendering component, thus improving its accuracy. This atomic rendering component can be obtained by encapsulating the rendering component in the first subroutine.

[0108] In one embodiment, during the execution of the first mini-program, the agent sends heartbeat information to the first mini-program; if no response to the heartbeat information is received within a preset time period, a third mini-program that provides business tasks for the first session message is selected. The third mini-program is invoked and executed to generate a fifth session message to reply to the first session message. This avoids the target atomic interface from being blocked indefinitely during execution, which could cause the agent process to be interrupted.

[0109] The main security runtime environment can refer to the framework's main context, the first-level sub-security runtime environment can refer to the business execution context, and the second-level security runtime environment can refer to the rendering context. The first-level sub-security runtime environment, the second-level sub-security runtime environment, and the main security runtime environment are all runtime environments provided by the host application for the first mini-program.

[0110] AppMainContext (Framework Main Context): Runs the framework code, manages the lifecycle, candidate atomic interface registration and scheduling, and message distribution. It has the highest privileges and runs on the host application's Biz thread.

[0111] AppSubContext (Business Execution Context): Runs developer code and creates an independent sandbox environment for each execution of the target atomic interface. It provides limited API access capabilities (a subset of wx.agentKit), has data acquisition capabilities such as network requests, and can be recycled after the target atomic interface is executed.

[0112] AppRenderContext (Rendering Context): Runs developer code and manages the rendering of atomic rendering components. It renders structuredContent into GUI cards through the magicBrush dynamic card mechanism. It does not have network request capabilities, and the rendering data can only come from the results returned by the target atomic interface in AppSubContext.

[0113] In this design, the security isolation framework code (AppMainContext) and the developer code (AppSubContext, AppRenderContext) run in different JSContexts. The developer code is isolated within a restricted permission domain and cannot access the framework's high-privilege interfaces. Although AppSubContext and AppRenderContext are both developer code, they belong to different permission sets: the business execution domain allows network requests to obtain data, while the rendering domain prohibits network requests. Data can only be passed in through structured results returned by atomic interfaces, preventing the rendering layer from directly initiating uncontrolled network requests.

[0114] This application enables the display of real-time dynamic atomic interfaces, meaning that these atomic interfaces can initiate network requests to retrieve the latest state from the developer's server and update the interface. Similarly, these atomic interfaces have different permission sets, and through domain isolation, AppRenderContext2 is added to render these atomic components.

[0115] Context lifecycle and card recovery: Creation: AppMainContext is created when a user enters the Agent conversation; Mounting: Loads the mini-program's mcp.json configuration and registers the declarations of candidate atomic interfaces; Execution: Create AppSubContext as needed and execute the target atomic interface call; Recycling: A lazy recycling strategy is used, which is triggered only when memory pressure is high; Card recovery: When the Context is recycled, the rendered message cards are displayed in a downgraded manner through screenshots (preserving visual snapshots). When the user interacts again, a cold start recovery is triggered (the Context is recreated and its state is restored), achieving memory optimization that is "unnoticeable to the user".

[0116] S205. The second session message is displayed on the session page.

[0117] In this application, when the second session message is generated, it can be displayed on the session page to enable interaction with the user and perform the first business task for the user.

[0118] It should be noted that during the runtime of the intelligent agent, the communication between the first mini-program and the intelligent agent adopts the private protocol of the host application, which is uniformly forwarded by the host application. The host application can refer to the application that provides the runtime environment for the atomic interface and atomic rendering component.

[0119] For example, the message types in the uplink channel that sends data from the platform backend to the host application include tool_call and heartbeat. tool_call is a call instruction (including interface name and input parameters) initiated by the platform backend regarding the target atomic interface, while heartbeat is a heartbeat message used for anti-freeze detection.

[0120] For example, the message types in the downlink channel where the host application sends data to the platform backend include tool_result, error, and context_event; tool_result is the execution result of the target atomic interface returned, error is the execution error information; context_event is the Context lifecycle event (creation / mounting / destruction).

[0121] In this application, all communication is forwarded by the host application, and developers do not need to access the communication link. The communication reuses the existing secure channel of the host application, which naturally supports encryption and authentication. User authentication reuses wx.login instead of the standard MCP OAuth 2.1 process, reducing the cost for smart agents to call the mini-program.

[0122] In this application, the first mini-program is invoked by the intelligent agent to execute the first business task indicated by the first session message, so as to generate a second session message for replying to the first session message. This eliminates the need for the user to manually select the mini-program to run after multiple attempts, improves the invocation efficiency of the mini-program, and provides users with more efficient and convenient business tasks.

[0123] Please see Figure 11 , Figure 11 This is a schematic diagram of the structure of a data processing device provided in this application. Figure 11 As shown, the data processing apparatus may include: The first display module 1011 is used to display the first conversation message input by the conversation object in the conversation page corresponding to the intelligent agent; The second display module 1012 is used to display a second session message in the session page for replying to the first session message; the second session message is generated by the agent calling the first applet to execute the first business task, and the first business task is the business task indicated by the first session message.

[0124] The second display module 1012 is used to display the conversation page corresponding to the smart agent in response to a trigger operation on the smart mini-program associated with the smart agent.

[0125] The second display module 1012 is used to display a call prompt message about the first mini-program on the session page; the call prompt message is used to reflect that the intelligent agent is calling the first mini-program. The second display module 1012 is also used to display execution progress information about the first mini-program on the session page; the execution progress information is used to reflect the progress of the first mini-program in executing the first business task.

[0126] The second display module 1012 is used to display, on the session page, a third session message entered by the session object in response to the second session message; The conversation page displays a fourth conversation message used to reply to the third conversation message; the fourth conversation message is generated by the agent calling the second mini-program to execute a second business task based on context information, the second business task being the business task indicated by the third conversation message, and the context information including at least one of the first conversation message, the second conversation message, or the third conversation message.

[0127] The second display module 1012 is used to display the business page of the first mini-program in response to a trigger operation for the second session message; The business page displays the business content associated with the second session message.

[0128] The second display module 1012 is used to display authorization request information on the session page; the authorization request information is used to indicate whether to authorize the first applet to obtain the object attribute information of the session object. In response to the confirmation operation for the requested authorization information, a second session message for replying to the first session message is displayed on the session page; the second session message is generated by the agent calling the first applet to execute the first business task based on the object attribute information. Optionally, the second display module 1012 is used for: The message card corresponding to the first mini-program is displayed on the conversation page; The message card displays a second session message used to reply to the first session message.

[0129] The second display module 1012 is used to display an image on the session page when the invocation time of the first mini-program is greater than a time threshold; the image is obtained by taking a screenshot of a message card that includes the second session message.

[0130] Optionally, the second display module 1012 is configured to, in response to an interactive operation on the session page, restore the display of a message card including the second session message on the session page.

[0131] Optionally, the second display module 1012 is also used for: At least one candidate mini-program associated with the first conversation message is displayed on the conversation page; In response to the selection operation for any candidate mini-program among the at least one candidate mini-programs, the selected mini-program is determined as the first mini-program for performing the first business task.

[0132] Optionally, the second display module 1012 is also used for: During the process of calling the third mini-program to reply to the first conversation message, in response to the message input operation on the conversation page, the change message entered by the conversation object is displayed on the conversation page; the change message is used to indicate the change of mini-program; Re-select the first mini-program from the mini-program list to perform the first business task.

[0133] Optionally, the second display module 1012 is also used for: Display the control for calling the mini-program assistant in the first mini-program; In response to a trigger operation on the invoke control, the session page of the mini-program assistant is displayed; The first conversation message input by the conversation object is displayed on the conversation page corresponding to the intelligent agent, including: In response to an input operation on the conversation page, the first conversation message entered by the conversation object is displayed on the conversation page of the mini-program assistant.

[0134] In this application, a smart agent calls and executes a first mini-program to provide business tasks for the first session message, thereby generating a second session message to reply to the first session message. This eliminates the need for the user to manually select the mini-program to run after multiple attempts, improving the efficiency of mini-program invocation and providing users with more efficient and convenient business tasks.

[0135] Please see Figure 12 , Figure 12 This is a schematic diagram of the structure of a data processing device provided in this application. Figure 12 As shown, the data processing apparatus may include: The first display module 1111 is used to display the first conversation message input by the conversation object in the conversation page corresponding to the intelligent agent; The acquisition module 1112 is used to acquire the program description information of the candidate mini programs in the mini program list; The determining module 1113 is used to determine, through the intelligent agent, a first mini-program for performing a first business task from the mini-program list based on the program description information; the first business task is the business task indicated by the first session message. The module 1114 is invoked to invoke the first mini-program to execute the first business task through the intelligence, so as to generate a second session message for replying to the first session message; The second display module 1115 is used to obtain the functional description information of the candidate atomic interfaces of the first mini-program; and to determine the target atomic interface for executing the first business task from the candidate atomic interfaces of the first mini-program according to the functional description information. The intelligent agent invokes the target atomic interface to execute the first business task, thereby generating a second session message to reply to the first session message.

[0136] The calling module 1114 is used to generate a calling instruction for the target atomic interface through the intelligent agent, and send the calling instruction to the first-level sub-safety operating environment through the main safe operating environment; In the first-level sub-security operating environment, according to the invocation instruction, the target atomic interface is invoked to execute the first business task, so as to generate a second session message for replying to the first session message.

[0137] The module 1114 is used to invoke the target atomic interface through the intelligent agent to execute the first business task and obtain the execution result; Through the intelligent agent, the atomic rendering component of the first mini-program is invoked in the second-level sub-security operating environment; The execution result is rendered using the atomic rendering component to obtain a second session message used to reply to the first session message.

[0138] Optionally, module 1113 is determined, specifically for: The intelligent agent selects a candidate mini-program associated with the first conversation message from the mini-program list based on the program description information. When there are multiple candidate mini-programs selected, obtain the historical call count of the session object for each selected candidate mini-program; Based on the historical call count, a first mini-program is determined from the selected candidate mini-programs to perform the first business task.

[0139] The module 1114 is invoked to send heartbeat information to the first mini-program through the intelligent agent during the execution of the first mini-program. If no response is received for the heartbeat information within a preset time period, a third mini-program that provides business tasks for the first session message is selected. The third subprogram is invoked and executed to generate a fifth session message to reply to the first session message.

[0140] Optionally, the determining module 1113, through the intelligent agent, determines from the list of mini-programs the first mini-program that provides the business task for the first session message, based on the program description information, including: The intelligent agent selects a candidate mini-program associated with the first conversation message from the mini-program list based on the program description information. When there are multiple candidate mini-programs selected, obtain the historical call count of the session object for each selected candidate mini-program; Based on the historical call count, the first mini-program that provides the business task for the first session message is determined from the selected candidate mini-programs.

[0141] In this application, the first mini-program is invoked by the intelligent agent to execute the first business task indicated by the first session message, so as to generate a second session message for replying to the first session message. This eliminates the need for the user to manually select the mini-program to run after multiple attempts, improves the invocation efficiency of the mini-program, and provides users with more efficient and convenient business tasks.

[0142] Please see Figure 13 , Figure 13 This is a schematic diagram of the structure of a computer device provided in this application. For example... Figure 13 As shown, the aforementioned computer device 1000 can refer to a server or terminal, including: a processor 1001, a network interface 1004, and a memory 1005. Furthermore, the aforementioned computer device 1000 may also include: a user interface 1003, and at least one communication bus 1002. The communication bus 1002 is used to implement communication between these components. In some embodiments, the user interface 1003 may include a display screen and a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as at least one disk storage device. The memory 1005 may also optionally be at least one storage device located remotely from the aforementioned processor 1001. Figure 13 As shown, the memory 1005, which is a computer-readable storage medium, may include an operating system, a network communication module, a user interface module, and computer programs.

[0143] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: The first conversation message input by the conversation object is displayed on the conversation page corresponding to the intelligent agent; The conversation page displays a second conversation message used to reply to the first conversation message; the second conversation message is generated by the agent calling the first mini-program to execute the first business task, and the first business task is the business task indicated by the first conversation message.

[0144] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: In response to a trigger operation on a smart mini-program associated with a smart agent, the corresponding conversation page for the smart agent is displayed.

[0145] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: The session page displays a prompt message about the first mini-program; the prompt message is used to reflect that the agent is calling the first mini-program.

[0146] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: The session page displays execution progress information about the first mini-program; the execution progress information is used to reflect the progress of the first mini-program in executing the first business task.

[0147] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: The session page displays the third session message entered by the session object in response to the second session message; The conversation page displays a fourth conversation message used to reply to the third conversation message; the fourth conversation message is generated by the agent calling the second mini-program to execute a second business task based on context information, the second business task being the business task indicated by the third conversation message, and the context information including at least one of the first conversation message, the second conversation message, or the third conversation message.

[0148] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: In response to a triggered operation for the second session message, the business page of the first mini-program is displayed; The business page displays the business content associated with the second session message.

[0149] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: The authorization request information is displayed on the session page; the authorization request information is used to indicate whether to authorize the first mini-program to obtain the object attribute information of the session object; In response to the confirmation operation for the requested authorization information, a second session message for replying to the first session message is displayed on the session page; the second session message is generated by the agent calling the first mini-program to execute the first business task based on the object attribute information.

[0150] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: The message card corresponding to the first mini-program is displayed on the conversation page; The message card displays a second session message used to reply to the first session message.

[0151] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: When the invocation time of the first mini-program exceeds a time threshold, an image is displayed on the session page; the image is obtained by taking a screenshot of a message card that includes the second session message.

[0152] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: In response to an interactive operation on the session page, the message card including the second session message is restored to be displayed on the session page.

[0153] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: At least one candidate mini-program associated with the first conversation message is displayed on the conversation page; In response to the selection operation for any candidate mini-program among the at least one candidate mini-programs, the selected mini-program is determined as the first mini-program for performing the first business task.

[0154] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: During the process of calling the third mini-program to reply to the first conversation message, in response to the message input operation on the conversation page, the change message entered by the conversation object is displayed on the conversation page; the change message is used to indicate the change of mini-program; Re-select the first mini-program from the mini-program list to perform the first business task.

[0155] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: Display the control for calling the mini-program assistant in the first mini-program; In response to a trigger operation on the invoke control, the session page of the mini-program assistant is displayed; The first conversation message input by the conversation object is displayed on the conversation page corresponding to the intelligent agent, including: In response to an input operation on the conversation page, the first conversation message entered by the conversation object is displayed on the conversation page of the mini-program assistant.

[0156] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: The first conversation message input by the conversation object is displayed on the conversation page corresponding to the intelligent agent; Retrieve the program description information of candidate mini programs in the mini program list; The intelligent agent determines a first mini-program from the mini-program list to perform the first business task based on the program description information; the first business task is the business task indicated by the first session message. Through the aforementioned intelligence, the first mini-program is invoked to execute the first business task, thereby generating a second session message for replying to the first session message; The second session message is displayed on the session page.

[0157] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: Obtain the functional description information of the candidate atomic interfaces of the first mini-program; Based on the functional description information, the target atomic interface for executing the first business task is determined from the candidate atomic interfaces of the first mini-program. The intelligent agent invokes the target atomic interface to execute the first business task, thereby generating a second session message to reply to the first session message.

[0158] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: The intelligent agent generates a call instruction for the target atomic interface and sends it to the first-level sub-security environment through the main security environment. In the first-level sub-security operating environment, according to the invocation instruction, the target atomic interface is invoked to execute the first business task, so as to generate a second session message for replying to the first session message.

[0159] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: The intelligent agent invokes the target atomic interface to execute the first business task and obtain the execution result. Through the intelligent agent, the atomic rendering component of the first mini-program is invoked in the second-level sub-security operating environment; The execution result is rendered using the atomic rendering component to obtain a second session message used to reply to the first session message.

[0160] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: The intelligent agent selects a candidate mini-program associated with the first conversation message from the mini-program list based on the program description information. When there are multiple candidate mini-programs selected, obtain the historical call count of the session object for each selected candidate mini-program; Based on the historical call count, a first mini-program is determined from the selected candidate mini-programs to perform the first business task.

[0161] Optionally, the processor 1001 can be used to call a computer program stored in the memory 1005 to implement the following steps of this application: During the execution of the first mini-program, the intelligent agent sends heartbeat information to the first mini-program; If no response is received for the heartbeat information within a preset time period, a third mini-program is selected to execute the first business task; The third mini-program is invoked to execute the first business task, so as to generate a fifth session message for replying to the first session message.

[0162] In this application, the first mini-program is invoked by the intelligent agent to execute the first business task indicated by the first session message, so as to generate a second session message for replying to the first session message. This eliminates the need for the user to manually select the mini-program to run after multiple attempts, improves the invocation efficiency of the mini-program, and provides users with more efficient and convenient business tasks.

[0163] Furthermore, it should be noted that this application also provides a computer-readable storage medium storing a computer program executed by the aforementioned data processing apparatus. This computer program includes program instructions, which, when executed by the processor, enable the execution of the data processing method described in the corresponding embodiments above. Therefore, further details will not be repeated here. Additionally, the beneficial effects of using the same method will also not be repeated. For technical details not disclosed in the embodiments of the computer-readable storage medium involved in this application, please refer to the description of the method embodiments of this application.

[0164] As an example, the above program instructions can be deployed and executed on a computer device, or deployed and executed on at least two computer devices in one location, or executed on at least two computer devices distributed in at least two locations and interconnected by a communication network. At least two computer devices distributed in at least two locations and interconnected by a communication network can form a blockchain network.

[0165] The aforementioned computer-readable storage medium can be a data processing apparatus provided in any of the foregoing embodiments or a central storage unit of the aforementioned computer device, such as a hard disk or central storage of the computer device. The computer-readable storage medium can also be an external storage device of the computer device, such as a plug-in hard disk, smart memory card (SMC), secure digital (SD) card, flash card, etc., provided on the computer device. Furthermore, the computer-readable storage medium may include both the central storage unit and external storage devices of the computer device. The computer-readable storage medium is used to store the computer program and other programs and data required by the computer device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0166] The terms "first," "second," etc., in the specification, claims, and drawings of this application are used to distinguish content in different media, rather than to describe a specific order. Furthermore, the term "comprising," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or modules, but may optionally include steps or modules not listed, or may optionally include other step units inherent to these processes, methods, apparatuses, products, or devices.

[0167] In practice, the collection and processing of data in this application should strictly comply with the requirements of relevant laws and regulations, obtain the informed consent or separate consent of the data subject, and carry out subsequent data use and processing within the scope of laws and regulations and the authorization of the data subject.

[0168] This application also provides a computer program product, including a computer program. When executed by a processor, the computer program implements the data processing method and decoding method described in the preceding embodiments, and therefore will not be repeated here. Furthermore, the beneficial effects of using the same method will also not be repeated. For technical details not disclosed in the embodiments of the computer program product involved in this application, please refer to the description of the method embodiments of this application.

[0169] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0170] The methods and related apparatus provided in this application are described with reference to the method flowcharts and / or structural diagrams provided in this application. Specifically, each block of the method flowchart and / or structural diagram, as well as combinations of blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable network-connected device to create a machine, such that the instructions, which execute via the processor of the computer or other programmable network-connected device, generate instructions for implementing the process. Figure 1 A schematic diagram of one or more processes and / or structures. Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable network-connected device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 A schematic diagram of one or more processes and / or structures. Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable network-connected device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable device for implementing the process. Figure 1 A process or multiple processes and / or structures illustrate the steps of the functions specified in one or more boxes.

[0171] The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Therefore, any equivalent variations made in accordance with the claims of this application shall still fall within the scope of this application.

Claims

1. A data processing method, characterized in that, include: The first conversation message input by the conversation object is displayed on the conversation page corresponding to the intelligent agent; The conversation page displays a second conversation message used to reply to the first conversation message; The second session message is generated by the agent calling the first mini-program to execute the first business task, and the first business task is the business task indicated by the first session message.

2. The method according to claim 1, characterized in that, The method further includes: In response to a trigger operation on a smart mini-program associated with a smart agent, the corresponding conversation page for the smart agent is displayed.

3. The method according to claim 1, characterized in that, The method further includes: The session page displays a prompt message about the first mini-program; the prompt message is used to reflect that the agent is calling the first mini-program.

4. The method according to claim 1, characterized in that, The method further includes: The session page displays execution progress information about the first mini-program; the execution progress information is used to reflect the progress of the first mini-program in executing the first business task.

5. The method according to claim 1, characterized in that, The method further includes: The session page displays the third session message entered by the session object in response to the second session message; The conversation page displays a fourth conversation message used to reply to the third conversation message; the fourth conversation message is generated by the agent calling the second mini-program to execute a second business task based on context information, the second business task being the business task indicated by the third conversation message, and the context information including at least one of the first conversation message, the second conversation message, or the third conversation message.

6. The method according to claim 1, characterized in that, The method further includes: In response to a triggered operation for the second session message, the business page of the first mini-program is displayed; The business page displays the business content associated with the second session message.

7. The method according to claim 1, characterized in that, The second session message displayed on the session page as a reply to the first session message includes: The authorization request information is displayed on the session page; the authorization request information is used to indicate whether to authorize the first mini-program to obtain the object attribute information of the session object; In response to the confirmation operation for the requested authorization information, a second session message for replying to the first session message is displayed on the session page; the second session message is generated by the agent calling the first mini-program to execute the first business task based on the object attribute information.

8. The method according to claim 1, characterized in that, The second session message displayed on the session page as a reply to the first session message includes: The message card corresponding to the first mini-program is displayed on the conversation page; The message card displays a second session message used to reply to the first session message.

9. The method according to claim 1, characterized in that, The method further includes: When the invocation time of the first mini-program exceeds a time threshold, an image is displayed on the session page; the image is obtained by taking a screenshot of a message card that includes the second session message.

10. The method according to claim 9, characterized in that, The method further includes: In response to an interactive operation on the session page, the message card including the second session message is restored to be displayed on the session page.

11. The method according to claim 1, characterized in that, The method further includes: At least one candidate mini-program associated with the first conversation message is displayed on the conversation page; In response to the selection operation for any candidate mini-program among the at least one candidate mini-programs, the selected mini-program is determined as the first mini-program for performing the first business task.

12. The method according to claim 1, characterized in that, The method further includes: During the process of calling the third mini-program to reply to the first conversation message, in response to the message input operation on the conversation page, the change message entered by the conversation object is displayed on the conversation page; the change message is used to indicate the change of mini-program; Re-select the first mini-program from the mini-program list to perform the first business task.

13. The method according to claim 1, characterized in that, The method further includes: Display the control for calling the mini-program assistant in the first mini-program; In response to a trigger operation on the invoke control, the session page of the mini-program assistant is displayed; The first conversation message input by the conversation object is displayed on the conversation page corresponding to the intelligent agent, including: In response to an input operation on the conversation page, the first conversation message entered by the conversation object is displayed on the conversation page of the mini-program assistant.

14. A data processing method, characterized in that, include: The first conversation message input by the conversation object is displayed on the conversation page corresponding to the intelligent agent; Retrieve the program description information of candidate mini programs in the mini program list; The intelligent agent determines a first mini-program from the mini-program list to perform the first business task based on the program description information; the first business task is the business task indicated by the first session message. Through the aforementioned intelligence, the first mini-program is invoked to execute the first business task, thereby generating a second session message for replying to the first session message; The second session message is displayed on the session page.

15. The method according to claim 14, characterized in that, The step of using the intelligence to invoke the first mini-program to execute the first business task, in order to generate a second session message for replying to the first session message, includes: Obtain the functional description information of the candidate atomic interfaces of the first mini-program; Based on the functional description information, the target atomic interface for executing the first business task is determined from the candidate atomic interfaces of the first mini-program. The intelligent agent invokes the target atomic interface to execute the first business task, thereby generating a second session message to reply to the first session message.

16. The method according to claim 15, characterized in that, The step of invoking the target atomic interface through the intelligent agent to execute the first business task, in order to generate a second session message for replying to the first session message, includes: The intelligent agent generates a call instruction for the target atomic interface and sends the call instruction to the first-level sub-security environment through the main security environment. In the first-level sub-security operating environment, according to the invocation instruction, the target atomic interface is invoked to execute the first business task, so as to generate a second session message for replying to the first session message.

17. The method according to claim 15, characterized in that, The step of invoking the target atomic interface through the intelligent agent to execute the first business task, in order to generate a second session message for replying to the first session message, includes: The intelligent agent invokes the target atomic interface to execute the first business task and obtain the execution result. Through the intelligent agent, the atomic rendering component of the first mini-program is invoked in the second-level sub-security operating environment; The execution result is rendered using the atomic rendering component to obtain a second session message used to reply to the first session message.

18. The method according to claim 14, characterized in that, The step of determining a first mini-program for performing the first business task from the mini-program list through the intelligent agent based on the program description information includes: The intelligent agent selects a candidate mini-program associated with the first conversation message from the mini-program list based on the program description information. When there are multiple candidate mini-programs selected, obtain the historical call count of the session object for each selected candidate mini-program; Based on the historical call count, a first mini-program is determined from the selected candidate mini-programs to perform the first business task.

19. The method according to claim 14, characterized in that, The method further includes: During the execution of the first mini-program, the intelligent agent sends heartbeat information to the first mini-program; If no response is received for the heartbeat information within a preset time period, a third mini-program is selected to execute the first business task; The third mini-program is invoked to execute the first business task, so as to generate a fifth session message for replying to the first session message.

20. A data processing apparatus, characterized in that, include: The first display module is used to display the first conversation message input by the conversation object in the conversation page corresponding to the intelligent agent; The second display module is used to display a second session message in the session page as a reply to the first session message; The second session message is generated by the agent calling the first mini-program to execute the first business task, and the first business task is the business task indicated by the first session message.

21. A data processing apparatus, characterized in that, include: The first display module is used to display the first conversation message input by the conversation object in the conversation page corresponding to the intelligent agent; The acquisition module is used to obtain the program description information of candidate mini programs in the mini program list; The determining module is configured to, through the intelligent agent, determine, from the list of mini-programs a first mini-program for performing a first business task based on the program description information; the first business task is the business task indicated by the first session message. The calling module is used to invoke the first mini-program to execute the first business task through the intelligence, so as to generate a second session message for replying to the first session message; The second display module is used to display the second session message on the session page.

22. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 19.

23. A computer storage medium, characterized in that, The computer storage medium stores a computer program, which, when executed by a processor, performs the method as described in any one of claims 1 to 19.

24. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 19.

Images