Software interaction method and device based on intelligent agent and medium

Abstract

The invention relates to the field of artificial intelligence, and particularly discloses an agent-based software interaction method and device and a medium, and the method comprises the steps: obtaining a natural language instruction input by a user and pre-stored user information in response to an identity verification passing instruction; determining an execution instruction template group corresponding to the natural language instruction and the user information in a preset database; generating an execution instruction group based on the natural language instruction and the execution instruction template group; matching execution functions corresponding to the execution instructions in a local function library based on a preset mapping relation; and executing the execution functions in sequence to obtain an execution result corresponding to the natural language instruction. By pre-storing user information, the same instruction can dynamically map differentiated operation logic according to identities. And under the condition of agent decision making, the user only needs to confirm the decision making of the agent after inputting the fuzzy demand, so that the condition of manual intervention is reduced.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, specifically to a software interaction method, device, and medium based on intelligent agents. Background Technology

[0002] An agent is an entity capable of perceiving its environment, making autonomous decisions, and taking actions to achieve specific goals. When an agent assists a user in a task, the user inputs natural language commands through a conversational application. The large language model bound to the agent then parses the natural language commands, retrieving matching target software and operation types. Based on the software interface support, it selects either an API interface or an RPA (Robotic Process Automation) agent interface associated with the software as the execution channel.

[0003] Current intelligent agent systems based on natural language interaction face challenges in controlling local software, including differing execution directions for the same command across roles and cross-domain issues. Existing systems do not semantically link user roles with operational logic, preventing differentiated operations based on user roles. This limits interaction accuracy, necessitates user confirmation, prolongs response time, and increases the amount of information users need to provide. For cross-domain API calls, traditional solutions rely on server-side proxies or other forwarding tools, leading to high deployment complexity, increased response latency, and security vulnerabilities. Summary of the Invention

[0004] To address the aforementioned problems, this application proposes a software interaction method, device, and medium based on intelligent agents, wherein the method includes: In response to an authentication pass command, the system obtains the natural language command input by the user and pre-stored user information; determines the execution command template group corresponding to the natural language command and the user information in a preset database; generates an execution command group based on the natural language command and the execution command template group; matches the execution function corresponding to each execution command in the local function library based on a preset mapping relationship; and executes the execution function sequentially to obtain the execution result corresponding to the natural language command.

[0005] In one example, the user information includes at least one of user name, user position, preferred tools, and commonly used configuration parameters; the step of determining the natural language instruction and the execution instruction template group corresponding to the user information in the preset database specifically includes: determining the user information to be activated corresponding to the natural language instruction; and determining the execution instruction template group in the preset database based on the natural language instruction and the user information to be activated.

[0006] In one example, determining the execution instruction template group in a preset database based on the natural language instruction and the user information to be activated specifically includes: determining that no corresponding execution instruction template is matched in the preset database for the natural language instruction and the user information to be activated; generating a recommendation function interface based on the natural language instruction and the user information to be activated; and returning local interaction failure information and the recommendation function interface to the user.

[0007] In one example, generating an execution instruction group based on the natural language instruction and the execution instruction template group specifically includes: determining the target parameter type corresponding to multiple execution instruction templates; extracting the target parameter value corresponding to the target parameter type from the natural language instruction; filling the target parameter value into the execution instruction template based on the target parameter type to obtain the execution instruction; and determining the instruction order of each execution instruction in the execution instruction group based on the template order of each execution instruction template in the execution instruction template group.

[0008] In one example, before extracting the target parameter value corresponding to the target parameter type from the natural language instruction, the method further includes: determining whether the natural language instruction and the user information contain target parameter values ​​corresponding to all target parameter types; if the natural language instruction and the user information do not contain target parameter values ​​corresponding to all target parameter types, then determining the parameter type to be supplemented; based on the parameter type to be supplemented, obtaining the parameter value to be supplemented from the user, and incorporating the parameter value to be supplemented into the natural language instruction.

[0009] In one example, after determining the instruction order of each execution instruction in the execution instruction group, the method further includes: determining a first update content of user information based on the natural language instructions; determining a second update content of user information based on the instruction order of each execution instruction in the execution instruction group; determining a third update content of user information based on the target parameter value; and updating and storing the user information based on the first update content, the second update content, and the third update content.

[0010] In one example, the preset database contains multiple execution instruction templates, each corresponding to an execution function. The step of matching the execution function corresponding to each execution instruction in the local function library based on the preset mapping relationship specifically includes: determining the name of the execution instruction template corresponding to the execution instruction; determining the name of the execution function that has a mapping relationship with the name of the execution instruction template; and determining the execution function corresponding to the execution instruction based on the name of the execution function.

[0011] In one example, before determining the execution instruction template group corresponding to the natural language instruction and the user information in the preset database, the method further includes: acquiring training sample data, the training sample data including natural language instruction samples and user information samples; training the agent based on the training sample data so that the agent can determine the execution instruction template group corresponding to the natural language instruction and the user information in the preset database.

[0012] This application also provides a software interaction device based on an intelligent agent, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform: in response to an authentication pass instruction, acquiring a natural language instruction input by a user and pre-stored user information; determining an execution instruction template group corresponding to the natural language instruction and the user information in a preset database; generating an execution instruction group based on the natural language instruction and the execution instruction template group; matching the execution function corresponding to each execution instruction in a local function library based on a preset mapping relationship; and sequentially executing the execution function to obtain the execution result corresponding to the natural language instruction.

[0013] This application also provides a non-volatile computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to: in response to an authentication pass instruction, obtain a natural language instruction input by a user and pre-stored user information; determine an execution instruction template group corresponding to the natural language instruction and the user information in a preset database; generate an execution instruction group based on the natural language instruction and the execution instruction template group; match the execution function corresponding to each execution instruction in a local function library based on a preset mapping relationship; and execute the execution function sequentially to obtain the execution result corresponding to the natural language instruction.

[0014] The method proposed in this application can bring the following beneficial effects: 1. By pre-stored user information, the same command can dynamically map differentiated operation logic based on user identity, reducing manual user intervention. User information comes from a user database, consisting of existing basic user information and information extracted from historical user input. Users do not need to fill in information beforehand; even if the information is incomplete, it will be provided by the user through question-and-answer processes during task execution, eliminating the need for additional user input for subsequent uses. In agent-based decision-making, users only need to confirm the agent's decision after inputting fuzzy requirements, reducing manual intervention.

[0015] 2. Functions within the local software can be called directly through predefined mapping relationships without relying on server-side request forwarding, thus avoiding security risks such as data leakage and interface exposure that may be caused by the existence of a proxy layer. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart illustrating a software interaction method based on an intelligent agent, as described in an embodiment of this application. Figure 2 This is a schematic diagram of the structure of a software interaction device based on an intelligent agent according to an embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0019] Figure 1 This diagram illustrates a process flow of an agent-based software interaction method provided in one or more embodiments of this specification. This method can be applied to various business domains, such as internet finance, e-commerce, instant messaging, gaming, and government services. The process can be executed by an agent pre-installed in a computing device within the corresponding domain. Certain input parameters or intermediate results within the process can be manually adjusted to improve accuracy.

[0020] The analysis method involved in the embodiments of this application can be implemented by a terminal device or a server. The server can be a single device or a system composed of multiple devices, i.e., a distributed server. This application does not specifically limit it in this way. For ease of understanding and description, the following embodiments are all described in detail using an intelligent agent as an example.

[0021] like Figure 1 As shown, this application provides a software interaction method based on an intelligent agent, including: S101: In response to the authentication pass instruction, obtain the natural language instruction entered by the user and the pre-stored user information.

[0022] When a user logs into a smart agent or smart application, identity verification is required. If the verification is successful, the smart agent can determine the pre-stored user information corresponding to that user based on the identity verification information (such as account name) entered by the user. This user information can be provided by the user or obtained by the smart agent from the user's historical commands.

[0023] Specifically, user information can be constructed as a user role vector R, which contains user information including user name, user position, preferred tools, and commonly used configuration parameters, such as {name, position, preferred tools, commonly used configuration parameters, ...}. If user information is not obtained before obtaining natural language commands, the corresponding vector elements will be empty. "Preferred tools" is a fixed-length array of commands, with the first element representing the most recently used command, and so on, meaning that commands listed earlier have higher priority.

[0024] S102: Determine the execution instruction template group corresponding to the natural language instruction and the user information in the preset database.

[0025] Since the content of natural language commands issued by users may be ambiguous, user information can be used to complete the commands, making them more comprehensive. Specifically, semantic information corresponding to the natural language command can be obtained through natural language processing technology. Then, based on this semantic information and pre-stored user information, a set of execution command templates corresponding to the user's natural language command is determined from a pre-defined database. Each execution command template set contains one or more execution command templates. When a user's natural language command can be completed with only one execution command, the execution command template set contains only one execution command template. When a user's natural language command requires multiple execution commands to complete, the execution command template set contains multiple execution command templates.

[0026] This execution instruction template is a standardized instruction with no parameters. Only the parameters need to be filled in to use it as an execution instruction for different execution functions. Each execution instruction template stores a description of the required parameters, forming an instruction library in tabular form. Text processing is performed by embedding it into the model, allowing large models to call instructions by reading from the knowledge base. Examples of some execution instruction template formats are shown below:

[0027] In one embodiment, when determining the execution instruction template group based on natural language instructions and user information, it is necessary to determine the user information to be activated corresponding to the natural language instructions; here, the user information to be activated refers to the user information required to execute the natural language instructions. Then, the execution instruction template group can be determined from a preset database based on the natural language instructions and the user information to be activated.

[0028] In one embodiment, the process of matching the execution instruction template described above can be performed by a trained agent. During training, the agent needs to be trained to determine the user information features to be activated from the natural language instructions provided by the user, and then match the corresponding execution instruction template. The requirement task corresponding to the natural language instruction can be represented as U, and the role vector corresponding to the user information can be represented as R. Here, the user information to be activated refers to the elements in R that need to be referenced to execute U. During the training phase, all information throughout the process is completely fed to the agent for learning. The fed role vector includes both complete and incomplete information, enabling the agent to react to situations where information is missing.

[0029] For example, when U is creating a task, the information that needs to be activated for U includes job positions, preferred instructions, and commonly used configurations in R. If the job position information in R is missing at this time, the user information to be activated includes preferred instructions (such as connecting to MATLAB, entering development mode, etc.). Since MATLAB is irrelevant to task creation, it can be assumed that the user prefers to create a development task, and finally, the user is directed to create a development task, prioritizing the acquisition of configuration parameters from commonly used configurations. When U has a sufficiently clear direction, the relevant features in R are not activated, and the user can directly be directed to create an office task. By feeding examples, the agent can learn which user information should be activated for a certain requirement, the user information to be prioritized, and the impact of user information on the vague requirements put forward by the user. After determining the user information to be activated corresponding to the natural language instructions and pointing to specific tasks based on the user information to be activated, the task can be decomposed to obtain different execution instruction templates. For example, creating a development task can be decomposed into an instruction to enter development mode and an instruction to create the task, which corresponds to two execution instruction templates.

[0030] In one embodiment, if no corresponding execution instruction template is found in the preset database for the natural language instruction and the user information to be activated, a recommended function interface is generated based on the natural language instruction and the user information to be activated; a local interaction failure message and the recommended function interface are returned to the user. Specifically, if there is no execution instruction template corresponding to the function to be implemented in the preset database, the user is informed that the function is not currently supported, and suggestions are provided to the user based on the user's own capabilities.

[0031] S103: Generate an execution instruction group based on the natural language instructions and the execution instruction template group.

[0032] After obtaining the execution instruction template group, parameters need to be filled into the execution instruction template group to obtain the execution instruction set. The execution instruction set can contain one or more execution instructions, and the number of execution instructions is the same as the number of execution instruction templates in the execution instruction template group.

[0033] In one embodiment, when generating an execution instruction group, since parameters need to be filled into the execution instruction template group, it is necessary to determine the target parameter types corresponding to various execution instruction templates. One execution instruction template may require one or more target table parameters. The agent then prioritizes obtaining the user's commonly used configuration parameters from user information, or extracts the target parameter values ​​corresponding to the target parameter types from natural language instructions and user information. For the task and its configuration parameters generated by the agent's decision, the agent displays them to the user and requests confirmation (the user replies "confirm" to the agent to confirm), ensuring the accuracy of the final task execution. Finally, based on the target parameter type, the target parameter values ​​are filled into the execution instruction template to obtain the execution instructions. The execution order of each execution instruction in the execution instruction group corresponds to the template order of each execution instruction template in the execution instruction template group.

[0034] In one embodiment, when generating the execution instruction group, parameters need to be filled into the execution instruction template group. However, the natural language instructions provided by the user may be too vague, resulting in incomplete required parameters. Therefore, before extracting the target parameter values ​​corresponding to the target parameter types from the natural language instructions, it is necessary to determine whether the natural language instructions and user information contain target parameter values ​​corresponding to all target parameter types. If the natural language instructions and user information do not contain target parameter values ​​corresponding to all target parameter types, then the parameter types to be supplemented are determined. Finally, based on the parameter types to be supplemented, the parameter values ​​to be supplemented can be obtained from the user and incorporated into the natural language instructions.

[0035] In one embodiment, after the task is determined to be executed but before actual execution, the stored user information needs to be updated. Specifically, if the natural language command provided by the user contains content related to user information, the first update content of the user information can be determined based on the natural language command; and the second update content of the user information can be determined based on the command order of each execution command in the execution command group, wherein the second update content mainly corresponds to the preferred tools. Then, based on the target parameter value, the third update content of the user information is determined, which mainly corresponds to commonly used configuration parameters. When updating and storing user information using the above first, second, and third update contents, the basic information or configuration parameters of the user in the first and third update contents can be directly replaced and updated; for the preferred tools in the second update content, the existing commands in the array can be moved to the beginning of the array; or a new command can be added to the beginning and the commands at the end of the array can be deleted.

[0036] S104: Match the execution function corresponding to each execution instruction in the local function library based on the preset mapping relationship.

[0037] After obtaining the execution instruction group, it is necessary to determine the corresponding execution function for each different execution instruction in the execution instruction group. Therefore, it is necessary to match the execution function corresponding to each execution instruction in the local function library based on the preset mapping relationship. The above execution functions are all stored in the local function library. The functions inside the local software can be called directly through the predefined mapping relationship without relying on the server to forward the request, thus avoiding security risks such as data leakage and interface exposure that may be caused by the existence of the proxy layer.

[0038] In one embodiment, a pre-stored database contains multiple execution instruction templates, and each instruction template corresponds to an execution function. When matching the execution function corresponding to each execution instruction, the name of the execution instruction template corresponding to the execution instruction can be determined first, and then the name of the execution function that has a mapping relationship with the name of the execution instruction template can be determined. Thus, based on the name of the execution function, the execution function corresponding to the execution instruction can be determined.

[0039] S105: Execute the execution functions sequentially to obtain the execution results corresponding to the natural language instructions.

[0040] After determining the execution function corresponding to the instruction group, the corresponding execution function can be executed sequentially based on the instruction order of the instructions in the instruction group, thereby obtaining the software execution result corresponding to the natural language instructions issued by the user.

[0041] At the user role level, this solution injects user identity data into agent training, enabling the agent to personalize the matching of execution instruction templates based on the user's identity. For cross-domain invocation issues (calling third-party tools or non-local interfaces), this solution can force the agent to output matching execution instructions based on the instruction template library, ensuring that local software can be directly mapped to the execution function without configuring proxies or installing plugins, thereby reducing deployment complexity.

[0042] like Figure 2 As shown, this application embodiment also provides a software interaction device based on an intelligent agent, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: In response to an authentication pass command, the system obtains the natural language command input by the user and pre-stored user information; determines the execution command template group corresponding to the natural language command and the user information in a preset database; generates an execution command group based on the natural language command and the execution command template group; matches the execution function corresponding to each execution command in the local function library based on a preset mapping relationship; and executes the execution function sequentially to obtain the execution result corresponding to the natural language command.

[0043] This application embodiment also provides a non-volatile computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured as follows: In response to an authentication pass command, the system obtains the natural language command input by the user and pre-stored user information; determines the execution command template group corresponding to the natural language command and the user information in a preset database; generates an execution command group based on the natural language command and the execution command template group; matches the execution function corresponding to each execution command in the local function library based on a preset mapping relationship; and executes the execution function sequentially to obtain the execution result corresponding to the natural language command.

[0044] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and medium embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the description of the method embodiments.

[0045] The devices and media provided in this application are one-to-one with the methods. Therefore, the devices and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0046] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0047] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, 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 data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0048] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing 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 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0049] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0050] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0051] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0052] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0053] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0054] The above description is merely an embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.

Claims

1. A software interaction method based on intelligent agents, characterized in that, include: In response to the authentication pass command, obtain the natural language command input by the user and the pre-stored user information; Determine the execution instruction template group corresponding to the natural language instruction and the user information in the preset database; Based on the natural language instructions and the execution instruction template group, an execution instruction group is generated; Based on a pre-defined mapping relationship, match the execution function corresponding to each execution instruction in the local function library; The execution functions are executed sequentially to obtain the execution results corresponding to the natural language instructions.

2. The method according to claim 1, characterized in that, The user information includes at least one of the following: user name, user job title, preferred tools, and commonly used configuration parameters. The step of determining the execution instruction template group corresponding to the natural language instruction and the user information in the preset database specifically includes: Determine the user information to be activated corresponding to the natural language command; Based on the natural language instructions and the user information to be activated, an execution instruction template group is determined in a preset database.

3. The method according to claim 2, characterized in that, The step of determining an execution instruction template group in a preset database based on the natural language instructions and the user information to be activated specifically includes: It is determined that the natural language instruction and the user information to be activated do not match a corresponding execution instruction template in the preset database; A recommendation function interface is generated based on the natural language instructions and the user information to be activated. Return local interaction failure information and the recommended function interface to the user.

4. The method according to claim 1, characterized in that, The step of generating an execution instruction group based on the natural language instructions and the execution instruction template group specifically includes: Determine the target parameter types corresponding to various execution instruction templates; Extract the target parameter value corresponding to the target parameter type from the natural language instruction; Based on the target parameter type, the target parameter value is filled into the execution instruction template to obtain the execution instruction; Based on the template order of each execution instruction template in the execution instruction template group, the instruction order of each execution instruction in the execution instruction group is determined.

5. The method according to claim 4, characterized in that, Before extracting the target parameter value corresponding to the target parameter type from the natural language instruction, the method further includes: Determine whether the natural language instruction and the user information contain target parameter values ​​corresponding to all target parameter types; If the natural language instructions and the user information do not contain all the target parameter values ​​corresponding to all target parameter types, then the parameter types to be supplemented are determined. Based on the type of parameter to be supplemented, the value of the parameter to be supplemented is obtained from the user, and the value of the parameter to be supplemented is incorporated into the natural language instruction.

6. The method according to claim 4, characterized in that, After determining the instruction order of each instruction in the instruction group, the method further includes: The first update content of the user information is determined based on the natural language instructions; Based on the instruction order of each execution instruction in the execution instruction group, the second update content of the user information is determined; Based on the target parameter value, determine the third update content of the user information; The user information is updated and stored based on the first update content, the second update content, and the third update content.

7. The method according to claim 1, characterized in that, The preset database contains a variety of execution instruction templates, and each instruction template corresponds to an execution function; The process of matching the execution function corresponding to each execution instruction in the local function library based on a preset mapping relationship specifically includes: Determine the name of the execution instruction template corresponding to the execution instruction; Determine the name of the execution function that has a mapping relationship with the name of the execution instruction template; Based on the execution function name, determine the execution function corresponding to the execution instruction.

8. The method according to claim 1, characterized in that, Before determining the execution instruction template group corresponding to the natural language instruction and the user information in the preset database, the method further includes: Acquire training sample data, which includes natural language instruction samples and user information samples; The agent is trained based on the training sample data so that it can determine the execution instruction template group corresponding to the natural language instruction and the user information in a preset database.

9. A software interaction device based on an intelligent agent, characterized in that, include: At least one processor; And, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform: In response to the authentication pass command, obtain the natural language command input by the user and the pre-stored user information; Determine the execution instruction template group corresponding to the natural language instruction and the user information in the preset database; Based on the natural language instructions and the execution instruction template group, an execution instruction group is generated; Based on a pre-defined mapping relationship, match the execution function corresponding to each execution instruction in the local function library; The execution functions are executed sequentially to obtain the execution results corresponding to the natural language instructions.

10. A non-volatile computer storage medium storing computer-executable instructions, characterized in that, The computer-executable instructions are set as follows: In response to the authentication pass command, obtain the natural language command input by the user and the pre-stored user information; Determine the execution instruction template group corresponding to the natural language instruction and the user information in the preset database; Based on the natural language instructions and the execution instruction template group, an execution instruction group is generated; Based on a pre-defined mapping relationship, match the execution function corresponding to each execution instruction in the local function library; The execution functions are executed sequentially to obtain the execution results corresponding to the natural language instructions.

Images