Multi-agent based sap system call and data interaction method and related device

Abstract

The application relates to the technical field of enterprise management systems, and in particular relates to a method for SAP system calling and data interaction based on multiple intelligent agents and related equipment, which obtains natural language information sent by a target object and a preconfigured routing mapping table, determines a target intelligent agent instance according to the natural language information, and looks up the authority record of the target object in the routing mapping table; if the target intelligent agent instance has a corresponding record in the authority record, the natural language information is input into the target intelligent agent instance to generate an interface function module; the interface function module corresponding to an interface credential is loaded in a preconfigured connection pool; the interface credential is sent to an SAP system, and after the SAP system returns authentication passing information, the interface function module is sent to the SAP system. The application can realize the use of natural language to request and call the SAP system, reduce the risk of unauthorized access, and be beneficial to maintaining data security.

Description

Technical Field

[0001] This application relates to the field of enterprise management system technology, and in particular to SAP system call and data interaction methods and related equipment based on multi-agent systems. Background Technology

[0002] SAP (Systems Applications and Products in Data Processing) is an enterprise resource planning software system. Traditionally, users operate the SAP system by entering transaction codes through a client-side GUI interface; one transaction code corresponds to one business operation. Given the large number of functions in the SAP system, users need to memorize a large number of transaction codes.

[0003] In related technologies, there is a solution that uses an agent to interact with an SAP system. This involves the agent recognizing the user's intent, translating that intent into an RFC function, connecting to the SAP system, and executing the RFC function to perform data processing. However, in practical applications, connecting to the SAP system via an agent has been found to pose risks of unauthorized access and data leakage. In current practices, users configure an RFC service account with broad privileges for the agent. Malicious users could potentially mislead the agent to access data they do not have permission to view, leading to data leakage. Summary of the Invention

[0004] In view of this, the main objective of the embodiments of this application is to propose a method and related device for SAP system invocation and data interaction based on multi-agent technology, which can realize the use of natural language to make requests to the SAP system, reduce the risk of unauthorized access, and help maintain data security.

[0005] To achieve the above objectives, one aspect of this application proposes a method for SAP system invocation and data interaction based on multi-agent systems, the method comprising: Obtain the natural language information sent by the target object and the pre-configured routing mapping table, determine the target intelligent agent instance based on the natural language information, and look up the permission record of the target object in the routing mapping table; If the target agent instance has a corresponding record in the permission record, the natural language information is input into the target agent instance to generate an interface function module; Load the interface credentials corresponding to the interface function module into the pre-configured connection pool; The interface credentials are sent to the SAP system, and after the SAP system returns authentication success information, the interface function module is sent to the SAP system.

[0006] In some embodiments, the method further includes: Obtain the configuration parameters of several intelligent agent instances, and configure several connection pools corresponding to each of the several intelligent agent instances according to the configuration parameters; Before sending the interface credentials to the SAP system, a connection object is created or returned from the connection pool corresponding to the target agent instance; Establish a session with the SAP system based on the connection object, and update the transaction context of the session.

[0007] In some embodiments, the step of inputting the natural language information into the target intelligent agent instance to generate the interface function module includes: The target agent instance is used to automatically scan the trigger scenario descriptions of each skill in the loaded skill set during the inference process; The target intelligent agent instance is used to determine whether the skill with the highest matching degree is greater than a preset matching degree threshold. If so, the skill with the highest matching degree is loaded, and the interface function module is generated based on the natural language information. If not, initiate a dialogue with the target object to obtain new natural language information.

[0008] In some embodiments, the skill further includes a processing flow, which includes parsing rules, verification rules, and function names; The process of loading the skill with the highest matching degree and generating the interface function module based on the natural language information includes: The parsing rules are determined, and the natural language information is extracted according to the parsing rules through the target intelligent agent instance to obtain the input parameters; Determine the verification rules, and perform integrity and format verification on the input parameters according to the verification rules using a target intelligent agent instance; Upon successful verification, the function name is determined, and the interface function module is generated based on the function name in the processing flow and the input parameters.

[0009] In some embodiments, the step of performing completeness and format checks on the input parameters according to the verification rules using the target agent instance includes: The missing parameters are determined according to the verification rules. The next round of dialogue is generated based on the missing parameters. The natural language information is updated. The execution steps are returned: the parsing rules are determined, and the natural language information is extracted by the target intelligent agent instance according to the parsing rules to obtain the input parameters.

[0010] In some embodiments, the method further includes: When the business scenario is updated, the dictionary specification corresponding to the transaction code is obtained, the processing flow and the triggering scenario description of the new skill are determined according to the dictionary specification, and the new skill is generated through the processing flow and the triggering scenario description.

[0011] In some embodiments, the method further includes: When the interface definition of the SAP system is updated, the new function name is obtained, and the old function name in the skill is replaced by the new function name.

[0012] To achieve the above objectives, another aspect of this application proposes a multi-agent-based SAP system call and data interaction device, the device comprising: The agent determination module is used to obtain natural language information sent by the target object and a pre-configured routing mapping table, determine the target agent instance based on the natural language information, and search for the permission record of the target object in the routing mapping table; The permission confirmation and request generation module is used to input the natural language information into the target intelligent agent instance and generate an interface function module if the target intelligent agent instance has a corresponding record in the permission record. The credential generation module is used to load the interface credential corresponding to the interface function module from the pre-configured connection pool; The calling module is used to send the interface credentials to the SAP system, and after the SAP system returns authentication success information, it sends the interface function module to the SAP system.

[0013] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0014] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned method.

[0015] To achieve the above objectives, another aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the aforementioned method. The embodiments of this application include at least the following beneficial effects: This application provides a method and related equipment for SAP system calls and data interaction based on multi-agent systems. This scheme obtains natural language information sent by a target object and a pre-configured routing table. Based on the natural language information, a target agent instance is determined. The permission record of the target object is searched in the routing table. If a corresponding record exists for the target agent instance in the permission record, the natural language information is input into the target agent instance to generate an interface function module. The interface credentials corresponding to the interface function module are loaded into a pre-configured connection pool. The interface credentials are sent to the SAP system. After the SAP system returns authentication success information, the interface function module is sent to the SAP system. When a target object performs business operations in the SAP system through a proxy agent instance, the target object can only access agent instances with access permission records. This can reject requests without the necessary access permissions for that agent instance. Furthermore, when a call request is made to the SAP system, the SAP system looks up the interface credentials in a table to verify whether the target agent instance has the right to execute the interface function module. This can reject unauthorized requests from malicious users that are injected through prompt words, reduce the risk of unauthorized access, and help maintain data security. Attached Figure Description

[0016] Figure 1 This is a flowchart of a method for SAP system invocation and data interaction based on multiple agents, provided in an embodiment of this application. Figure 2 This is a schematic diagram illustrating the steps for session management with the SAP system provided in an embodiment of this application; Figure 3 This is a schematic diagram illustrating the steps of matching skills from a skill set according to an embodiment of this application; Figure 4 This is a schematic diagram illustrating the steps of generating the interface function module provided in the embodiments of this application; Figure 5 This is a schematic diagram of the skill set provided in the embodiments of this application; Figure 6 This is a schematic diagram of the dialog interface for processing historical unprocessed daily settlement data reminders provided in the embodiments of this application; Figure 7 This is a schematic diagram of the dialogue interface for processing dialogue-driven product removal requests provided in this application embodiment; Figure 8 This is a schematic diagram of the structure of the SAP system call and data interaction device based on multiple agents provided in the embodiments of this application; Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0018] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0019] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0021] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0022] 1) SAP (Systems Applications and Products in Data Processing) is an enterprise resource planning software system used to manage enterprise data.

[0023] 2) SAP transaction codes are code identifiers used in the SAP system to quickly invoke specific functions. Users can directly enter the corresponding function interface by entering the transaction code in the command field of the SAP GUI and pressing Enter.

[0024] In the traditional approach, users operate the SAP system by entering transaction codes through the client's GUI interface, with one transaction code corresponding to one business operation. Given the large number of functions in the SAP system, users need to memorize a large number of transaction codes.

[0025] In related technologies, there is a solution that uses an agent to interact with an SAP system. This involves the agent recognizing the user's intent, translating that intent into an RFC function, connecting to the SAP system, and executing the RFC function to perform data processing. However, in practical applications, connecting to the SAP system via an agent has been found to pose risks of unauthorized access and data leakage. In current practices, users configure an RFC service account with broad privileges for the agent. Malicious users could potentially mislead the agent to access data they do not have permission to view, leading to data leakage.

[0026] Furthermore, SAP systems differ significantly from general database systems. SAP employs a transaction code system containing thousands of transaction codes, each corresponding to specific business operations and parameter requirements. Users need to memorize a large number of transaction codes and their operating rules. SAP systems have a complex access control system, including multiple layers such as user permissions, role permissions, transaction code permissions, and field-level permissions. SAP systems use the RFC (Remote Function Call) protocol for remote calls, requiring the maintenance of session state and transaction context, which is fundamentally different from general REST API calls. Data relationships between business modules in SAP systems are complex; a single business operation may involve data interaction across multiple modules / tables. Due to these differences, existing general-purpose models often misunderstand SAP interfaces when processing them, leading to incorrect interface call modules and call failures due to a lack of understanding of the strict constraints of the SAP dictionary (such as field length, format, and required fields).

[0027] Furthermore, SAP GUI automation operations implemented using RPA technology become ineffective when SAP GUI interface coordinates and control IDs are upgraded or the interface is fine-tuned, resulting in high maintenance costs; the hard-coded interface integration method is also difficult to adapt to SAP's frequently changing business processes.

[0028] In view of this, this application provides a method and related equipment for SAP system calls and data interaction based on multi-agent systems. This scheme obtains natural language information sent by the target object and a pre-configured routing table. Based on the natural language information, it determines the target agent instance and searches the routing table for the target object's permission record. If the target agent instance has a corresponding record in the permission record, it inputs the natural language information into the target agent instance to generate an interface function module. It loads the interface credentials corresponding to the interface function module into a pre-configured connection pool and sends the interface credentials to the SAP system. After the SAP system returns authentication success information, it sends the interface function module to the SAP system. When the target object performs business operations in the SAP system on behalf of an agent instance, the target object can only access agent instances with permission records. This rejects requests without the necessary access permissions. Furthermore, when initiating a call request to the SAP system, the SAP system looks up the interface credentials in a table to verify whether the target agent instance has the right to execute the interface function module. This rejects unauthorized requests from malicious users injected via prompt words, reducing the risk of unauthorized access and contributing to data security.

[0029] The SAP system call and data interaction method based on multi-agent provided in this application relates to the field of enterprise management system technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or in-vehicle terminal, but is not limited thereto. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the SAP system call and data interaction method based on multi-agent, but is not limited to the above forms.

[0030] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0031] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0032] Figure 1 This is an optional flowchart of the SAP system call and data interaction method based on multi-agent provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps 101 to 104.

[0033] Step 101: Obtain the natural language information sent by the target object and the pre-configured routing mapping table; determine the target intelligent agent instance based on the natural language information; and search for the permission record of the target object in the routing mapping table. The system predefines several agent instances, which can be categorized according to function, business domain, etc. For example, if the natural language information sent by the target object is: "What is the total amount of the accounts of subject 8000 as of August 2026?", then based on this information, it can be determined that the agent instance that processes financial data should handle the processing.

[0034] The routing table is used to represent the mapping relationship between target objects and agent instances. For example, target object 1 has mapping records with agent instance a, agent instance b, and agent instance c in the table, and target object 2 has a mapping record with agent instance d in the table.

[0035] Step 102: If the target agent instance has a corresponding record in the permission record, input the natural language information into the target agent instance to generate an interface function module; In step 102, if the current natural language information determines that the agent instance is a, then agent instance a will continue to process; if the current natural language information determines that the agent instance is d, and agent instance d has no mapping relationship with target object 1, then the message "target object does not have permission" will be returned.

[0036] Step 103: Load the interface credentials corresponding to the interface function module into the pre-configured connection pool; The interface credential is determined by a pre-configured function permission mapping table, which represents the mapping relationship between the interface credential and the interface function module.

[0037] For example, if the connection pool of the target intelligent agent instance contains pre-configured interface credentials L1, L2, and L3, and the currently generated interface function module has a mapping record with interface credentials L1 in the function permission mapping table, then interface credentials L1 will be loaded from the connection pool.

[0038] Interface credentials are distributed by the SAP system, which generates them according to the configured function permission mapping table. For example, after receiving interface credential L1, the SAP system queries the function permission mapping table and finds that the target agent instance does not have a mapping record for interface credential L1, then returns the message "Agent has no permission".

[0039] Step 104: Send the interface credentials to the SAP system. After the SAP system returns authentication success information, send the interface function module to the SAP system.

[0040] Steps 101 to 104, as illustrated in this embodiment, involve obtaining natural language information sent by the target object and a pre-configured routing table. The target agent instance is determined based on the natural language information, and the permission record of the target object is searched in the routing table. If a corresponding record exists for the target agent instance in the permission record, the natural language information is input into the target agent instance to generate an interface function module. The interface credential corresponding to the interface function module is loaded into a pre-configured connection pool, and the interface credential is sent to the SAP system. After the SAP system returns authentication success information, the interface function module is sent to the SAP system. When the target object performs business operations in the SAP system on behalf of the agent instance, the target object can only access agent instances with permission records. This rejects requests without the necessary permissions. Furthermore, when a call request is initiated to the SAP system, the SAP system checks the interface credential against the table to verify whether the target agent instance has the right to execute the interface function module. This rejects unauthorized requests from malicious users injected via prompt words, reducing the risk of unauthorized access and contributing to data security.

[0041] Furthermore, when a malicious user injects a hint word into a target agent instance to generate a fake interface credential, the SAP system can find by looking up a table that there is no mapping record between the target agent instance and the interface function module corresponding to the fake interface credential, and reject the request. As a result, the SAP system function call fails and returns "the agent has no authorization information".

[0042] See Figure 2 In some embodiments, the method further includes steps 201 to 203: Step 201: Obtain the configuration parameters of several intelligent agent instances, and configure several connection pools corresponding to each of the several intelligent agent instances according to the configuration parameters. The configuration parameters include the number of connections, the maximum number of idle connections, the maximum number of active connections, the longest waiting time, and the scan interval. It can be understood that a connection pool is a technical mechanism for managing and reusing backend service connections. During interaction with the SAP system, when the communication connection is used as a backend service for the SAP system, the frontend creates a connection pool according to the configuration parameters, establishing and maintaining several connections. When the target agent sends an interface function module, it is not necessary to re-establish the connection; instead, the existing connection is used for data interaction.

[0043] In step 201, each agent instance creates a connection pool, and the agent instance sends or receives data to or from the SAP system through the connection in its respective connection pool.

[0044] Step 202: Before sending the interface credentials to the SAP system, create or return a connection object from the connection pool corresponding to the target agent instance; The connection object, also known as the signal channel, is the connection mentioned above. It creates a new connection object from the connection pool or returns an existing and idle connection object. Data is sent to or received from the SAP system through the connection object.

[0045] Step 203: Establish a session with the SAP system based on the connection object, and update the transaction context of the session.

[0046] The transaction context is the session information of the current connection object. After selecting a connection object from the connection pool, the historical session information of that connection object is updated.

[0047] In this embodiment, steps 201 to 203 ensure that the instructions of different agent instances are represented as completely isolated login sessions in the underlying SAP system, thereby blocking cross-session memory cache pollution and data lock contention, which is beneficial for further permission management of interface function modules.

[0048] See Figure 3 In some embodiments, the step of inputting the natural language information into the target intelligent agent instance to generate the interface function module includes steps 301 to 303: Step 301: Obtain or update the natural language description; Step 302: During the inference process, the target agent instance automatically scans the trigger scenario descriptions of each skill in the loaded skill set. Specifically, the skill set includes several skills, each of which includes a triggering scenario description. The first feature vector is obtained by vectorizing the triggering scenario descriptions of the several skills.

[0049] Step 303: Use the target agent instance to determine whether the skill with the highest matching degree is greater than a preset matching degree threshold. If yes, load the skill with the highest matching degree and generate the interface function module based on the natural language information. If no, engage in dialogue with the target object to obtain new natural language information.

[0050] The target agent instance compares the natural language information input by the user with the loaded skill set using semantic vectors and performs contextual reasoning to calculate the intent matching degree. If the matching degree is lower than a preset threshold (e.g., 0.8), or if the target agent instance determines that the user's intent is ambiguous or ambiguous (e.g., only inputting "help me check for an anomaly"), the target agent instance will automatically generate a clarifying question (e.g., "Do you mean an anomaly in the daily financial settlement or an anomaly in the interface transmission?") and push the question to the chat interface interacting with the target object. After waiting for the target object to supplement the input, it will re-enter step 301.

[0051] In this embodiment, the application writes the trigger scene description into the skill, and selects the skill with the highest matching degree from the skill set based on the semantic similarity comparison result between the trigger scene description and the natural language information. This can improve the success rate of generating the interface function module.

[0052] See Figure 4 In some embodiments, the skill further includes a processing flow, which includes parsing rules, verification rules, and function names; loading the skill with the highest matching degree and generating the interface function module based on the natural language information includes: Step 401: Determine the parsing rule, and extract parameters from the natural language information using the target intelligent agent instance according to the parsing rule to obtain the input parameters; Specifically, the parsing rules are determined by the data types, data formats, and data structures specified in the SAP system.

[0053] For example, if the input message is "Date 20260501, 9 rows of unposted data; Date 20260502, 10 rows of unposted data; Date 20260503, 10 rows of unposted data", then the extracted result is: ["20260501", "20260502", "20260503"].

[0054] Step 402: Determine the verification rules, and perform integrity and format verification on the input parameters according to the verification rules through the target intelligent agent instance; In some specific embodiments of step 402, the target agent instance dynamically evaluates the completeness of the input parameters according to the verification rules. If it is determined that a required parameter is missing, the target agent instance generates a new round of dialogue based on the missing required parameter and returns to step 401. The required parameter is determined according to the input parameter of the function name specified in the SAP system.

[0055] For example: "No posting date detected. Please specify the date range."

[0056] In some specific embodiments of step 402, the extracted parameters are converted into strings that conform to the dictionary specification requirements of the SAP system.

[0057] Step 403: When the verification passes, determine the function name and generate the interface function module according to the function name of the processing flow and the input parameters.

[0058] In step 403, the input parameters and function name are assembled to obtain the interface function module. It is understood that the interface function module is a computer-executable instruction.

[0059] It should be noted that in the SAP system, interface functions have strict requirements on the format of assembly parameters. Directly generating interface functions from a general large model can easily lead to illusions, generating parameters that do not conform to the SAP system's dictionary specifications, resulting in request failures.

[0060] In this embodiment, the required fields, field types, and enumeration ranges of the input constraints are forcibly constrained through the parsing rules in the skill's processing flow. If the validation fails, the target agent instance uses multiple rounds of dialogue to question the target object about the missing or incorrect required parameters until all the input parameters meet the assembly requirements of the interface function module. This reduces the probability of irreversible business operations triggered by large model illusions.

[0061] In some embodiments, the method further includes: when the business scenario is updated, obtaining the dictionary specification corresponding to the transaction code, determining the processing flow and the triggering scenario description of the new skill according to the dictionary specification, and generating the new skill through the processing flow and the triggering scenario description.

[0062] It's important to note that SAP systems contain tens of thousands of transaction codes. A single module may have hundreds of commonly used transaction codes, and cross-module transactions can number in the thousands. Furthermore, memorizing these transaction codes relies primarily on internal operational experience and memory. When new transaction codes or functionalities need to be added to a business scenario, the transaction codes must be updated and encoded in the SAP system's GUI front-end.

[0063] In this embodiment, after the interface function module of the SAP system backend is defined, it is only necessary to extract the processing flow and the trigger scenario description from the dictionary specification of the interface function module to realize the skill update. In this way, only the code update of the SAP system needs to be completed, without the need to update the frontend synchronously.

[0064] In some embodiments, the method further includes: when the interface definition of the SAP system is updated, obtaining a new function name and replacing the old function name in the skill with the new function name.

[0065] It should be noted that SAP system interface function modules are updated frequently. Once the interface function modules of a skill that has been stored are updated, the old interface function modules cannot be requested.

[0066] In this embodiment, the SAP system integration is completed by updating the function name of the skill, without needing to update the front end synchronously.

[0067] The following is a detailed description and explanation of the solutions in the embodiments of the present invention, in conjunction with specific scenarios: This application provides a method for SAP system calls and data interaction based on multi-agent systems. This method is applied to agent tools that interact with the SAP system, replacing the SAP system's GUI front-end. SAP is an enterprise data management system. Users can navigate to corresponding functions by entering transaction codes in the GUI front-end and select the enterprise data to be added, deleted, modified, or queried according to the function menu. The agent tool identifies user intent through dialogue, converts the intent into interface module functions, and automatically executes the CRUD operations on enterprise data within the SAP system through these interface module functions. Currently, agents interacting with the SAP system are at risk of unauthorized access. The method in this application manages agent permissions.

[0068] See Figure 5 This application provides a method for SAP system calls and data interaction based on multiple agents, in which the agent instance is configured with a skill set containing multiple skills. Taking the skill name BU032-Historical Unprocessed Daily Settlement Data Reminder as an example, it includes the following structured content: (1) name say BU032 - Historical Unprocessed Daily Settlement Data Reminder (2) Triggering scenario description: > Process the BU032-historical unprocessed daily settlement data reminder pushed by Enterprise Robot, parse the unposted date in the message, and call SAPRFCZZF_intelligent agent instance_BU032 to obtain detailed unposted data information.

[0069] (3) Triggering scenario: This skill is triggered when a BU032 alarm message is received from the Enterprise Robot or submitted through the Console. The message format is as follows: "BU032 - Historical Unprocessed Daily Settlement Data Reminder" 1. Date 20260501, 9 rows of unposted data 2. Date 20260502, 10 rows of unposted data. 3. Date 20260503, 10 rows of unposted data. (4) Boundary description: This skill only performs query operations and does not modify any data. The date format is uniformly `YYYYMMDD` (8 digits). All operations are completed through the SAP RFC interface and support dual-channel input parameters: Enterprise WeChat push (with original message) or Console submission (without original message).

[0070] (5) Processing flow, including: Step 1: Parse the data, for example: "Analyze message extraction dates: Extract unposted dates from push messages and form a date array."

[0071] Parsing rules: 1. Extract an 8-digit number in the format `DateYYYYMMDD` from each row of data. 2. Ignore descriptive text such as line numbers and the number of unposted lines. 3. The date array is arranged in the order of the message, without repetition or sorting.

[0072] Step 2: Determine the function name, and generate the interface function module based on the function name of the processing flow and the input parameters.

[0073] For example: to retrieve unlisted data by calling SAP RFC, use the `call_rfc_function` tool to call the RFC function `ZZF_AGENT_BU032`.

[0074] Function name: `ZZF_AGENT_BU032`.

[0075] Validation rules: Validate missing and incorrect items in the input parameters according to the following table.

[0076] The input parameters are shown in Table 1.

[0077] Table 1: Input Parameters

[0078] The structure of the IT_BUDAT table is shown in Table 2.

[0079] Table 2: Structure of IT_BUDAT table

[0080] Step 3: Output the unlisted data details and organize the returned data into a table for output.

[0081] Example Scenario 1: Handling of Historical Unprocessed Daily Closing Data. After the close of business each day, the SAP retail operations module summarizes the daily sales figures by store and POS terminal, and posts the daily closing data to financial vouchers (such as SD-FI interfaces, sales vouchers, and inventory difference adjustment vouchers). Due to factors such as network interruptions, external POS upload delays, SAP backend lock conflicts, or update queue anomalies, a small amount of historical daily closing data that should have been posted but wasn't will remain in the intermediate tables. This data must be promptly checked and manually posted by operations personnel; otherwise, it will lead to distorted monthly closing data and difficulties in financial reconciliation. Traditionally, operations personnel need to log into SAPGUI, execute special transaction codes to access ALV reports, manually enter the date range and company code, export details to Excel, and then check each row. For multi-day data accumulated across multiple days, multiple queries need to be performed repeatedly. The entire process is time-consuming, relies on memorizing specialized transaction codes, and cannot be processed in a fragmented manner on mobile devices.

[0082] See Figure 6 The method provided in this application is applied in chatbots. The chatbot obtains natural language descriptions and triggers the background to execute the following steps to process historical unprocessed daily settlement data reminders: S1.1 Message Access and Identity Resolution: The instant messaging access layer receives alarm cards pushed by WeChat Webhook or follow-up messages from operations and maintenance personnel @SAP Smart Assistant. The entire message body (including the referenced alarm content and the current input content) is encapsulated in Base64 encoding as the IV_WX_INPUT field. At the same time, the WeChat UserId (e.g., zhangsan) and session ID of the message sender are extracted as the unique credentials for subsequent routing and auditing.

[0083] S1.2, Intelligent Routing and Intelligent Agent Instance Loading: The routing module retrieves the unique corresponding intelligent agent instance identifier (such as intelligent agent instance_SAP_OPS) in the user-intelligent agent instance binding table based on the user's WeChat account UserId, and loads the corresponding SAP RFC interface credentials (RFCUser, PFCG role, CPIC connection handle) in the credential pool based on the intelligent agent instance identifier. This binding relationship is written once during the system initialization phase and dynamic switching is not allowed during runtime.

[0084] S1.3, Business ID Identification and Skill Loading: The agent instance extracts entities from the message text based on the predefined prompt word template, identifies the predefined business identifier "BU032" or synonymous keywords such as "historical unprocessed daily settlement data reminder", and loads the corresponding BU032.md skill from the skill knowledge base.

[0085] S1.4, User Description Keyword Extraction: The agent instance jointly extracts the posting date entity from the referenced alarm card and the user's follow-up text, and parses the natural language "May 1st and May 2nd" into the SAP standard date array ["20260501", "20260502"]. If only part of the date is identified or the date range is ambiguous (such as "last Wednesday"), the agent instance generates clarification questions based on skill constraints and asks the user follow-up questions through multiple rounds of dialogue until the date parameters are complete and conform to the YYYYMMDD format.

[0086] S1.5, Protocol Adaptation and RFC Input Parameter Assembly: The protocol adaptation layer maps the standardized skill call request to the input parameters of the interface function module according to the BU032 skill: the original message is Base64 encoded as a whole and then passed to IV_WX_INPUT; the selection condition table IT_BUDAT is constructed row by row according to the date array, and each row is set to SIGN='I', OPTION='EQ', LOW=corresponding date; the interface credentials loaded in S1.2 are automatically passed through.

[0087] S1.6, RFC Call and Connection Reuse: The protocol adaptation layer reuses the established CPIC connection in the RFC connection pool exclusive to this agent instance, without repeating the Logon handshake, and calls the core query function ZZF_AGENT_BU032. The SAP backend retrieves the intermediate table based on the IT_BUDAT selection criteria, backfills the unposted details into the internal table, and returns the execution status in ES_AGENT_RETURN.

[0088] S1.7, Status Determination and Exception Handling: The agent instance parses the ES_AGENT_RETURN.TYPE field: If it returns 'S' (success), proceed to S1.8 result formatting; if it returns 'E' / 'A' (failure / abortion), further identify the error code—if it belongs to retryable exceptions such as RFC connection timeout or lock conflict, it will automatically retry according to the exponential backoff strategy (maximum 3 times); if it belongs to non-retryable exceptions such as permission denial or illegal parameters, or exceeds the retry limit, the error message will be mapped to business language and written back to WeChat and the process will end.

[0089] S1.8, Result Formatting and Dynamic Degradation Output: The agent instance cleans and maps fields (voucher number, posting date, company code, amount, reason for non-posting, person in charge) of the unposted details according to the output rules preset by BU032, and performs dynamic degradation based on the instant messaging display threshold (default 14 rows): when the data volume is less than the threshold, it is directly rendered as a Markdown structured table; when the data volume exceeds the threshold, the file generation module is automatically triggered, packaged as an Excel attachment and a download link is generated.

[0090] S1.9, Result Feedback and Closed Loop: The system asynchronously writes back to the chatbot via the original conversation channel, along with a structured table or Excel link and a summary of key indicators (N rows in total, distributed by date, with top responsible persons). Operations personnel can read and make decisions on the Enterprise WeChat side without logging into the SAP GUI, thus completing the closed loop of this natural language-driven processing flow.

[0091] This application achieves the following technical effects in the scenario of SAP historical unprocessed daily settlement data reminders: (1) Directly parse the semi-structured alarm cards pushed by the robot and the natural language follow-up questions from the maintenance personnel, and automatically complete the business number recognition (BU032), posting date array extraction and SAP standard date format conversion, IT_BUDAT selection condition table assembly and IV_WX_INPUTBase64 encoding, completely eliminating the tedious steps of manually parsing alarms and manually logging into SAPGUI to enter query conditions.

[0092] (2) The permission mapping table and routing table are statically bound during system initialization and cannot be changed during runtime. Every ZZF_AGENT_BU032 call triggered by the operation and maintenance personnel is executed in the SAPLogonSession exclusively owned by this instance. The PFCG role forcibly converges the permission scope at the SAP kernel level, fundamentally eliminating unauthorized access and blind spots in audit tracing caused by sharing high-privilege accounts.

[0093] (3) In view of the variable number of rows returned by RFC, the system has built-in dynamic routing and formatting logic based on the instant messaging display threshold: when the data volume is small, the Markdown text table is rendered directly; when the data volume is large, the Excel attachment is automatically generated and the download link is pushed, taking into account the display limitations of the message channel and the data integrity, and avoiding the chat interface from being overwhelmed.

[0094] (4) Reusing the RFC connection pool exclusively used by the agent instance avoids repeated CPIC handshakes, further improving the response speed and throughput in large-scale daily data screening scenarios.

[0095] Example Scenario 2: Processing Product Removal Requests Online platform third-party merchants' business operations staff need to initiate delisting applications for slow-moving, non-compliant, near-expiry, or discontinued products in their daily work. Traditionally, operations staff need to log in to the POP backend management system, click the "Product Management - Delisting Application - Create" button on the front-end interface, manually query the product's EAN code, warehouse, and delisting reason dictionary value (LOV), enter more than ten fields one by one, click "Save" and then click "Submit". The whole process is time-consuming, prone to omissions, and prone to selecting the wrong warehouse.

[0096] See Figure 7, processing product delisting requests through conversation driven by a chatbot, the method provided by the embodiments of the present application comprises the following steps: S2.1, Request for delisting application: the instant messaging access layer receives the original message body (Base64 encoded) pushed by WeChat Work Webhook, encapsulates metadata such as the user's WeChat Work UserId, message content, and session ID into a target object, and forwards it to the natural language processing and intention analysis layer.

[0097] S2.2, Intelligent routing and agent instance loading: retrieve the uniquely corresponding agent instance identifier (e.g., agent_instance_POP_OPS) in the user-agent instance binding table according to the user's WeChat Work UserId, and load the corresponding interface credential in the credential pool based on the agent instance identifier; the binding relationship among the three is written in one time during the system initialization stage, and dynamic switching is not allowed during runtime.

[0098] S2.3, Agent instance permission verification: the system adds two layers of permission verification at the agent instance level.

[0099] S2.4, Keyword extraction from user description. The agent instance performs entity extraction on the user input based on a predefined prompt template to obtain candidate keywords, for example, product name = "Nongfu Spring Drinking Water 550ml", warehouse name = "Guangzhou Central Warehouse", reason description = "unsalable", effective time = "next Monday".

[0100] S2.5, Input parameter completion: for strongly typed fields that cannot be directly given through natural language description, the agent instance calls an auxiliary tool for accurate conversion. The product name is parsed into eanCode through search_product_ean; "Guangzhou Central Warehouse" is parsed into dcCode / dcName through search_lov_warehouse; "unsalable" is parsed into sxjTypeCode and codeValue through search_lov_ean; "next Monday" is converted into effectiveDate (YYYY-MM-DD) through natural language time parsing. When any auxiliary query fails, the agent instance enters exponential backoff retry, and outputs failure information to the user and terminates when the maximum number of retries is exceeded.

[0101] S2.6, Strong Type Validation using JSON Schema. The agent instance validates the completed parameter object according to the pre-defined JSON Schema in the skill, forcibly constraining ten mandatory fields: eanCode, eanName, lv, dcCode, dcName, effectiveDate, sxjTypeCode, codeValue, sxjReason, and sxjDescription, specifying their types and enumeration ranges. If validation fails, the agent instance will prompt the user through multiple rounds of dialogue to inquire about missing or incorrect items until all fields meet the submission requirements.

[0102] S2.7, Manual Confirmation: Before calling any write operation MCP interface, the agent instance must echo the complete parameters back to WeChat Work in the form of a readable card and wait for the user to explicitly reply "confirm". This rule is written into the skill description, and calling the interface without confirmation is strictly prohibited.

[0103] In step S2.8, after user confirmation, the agent instance calls `pop_delisting_tool` with `action=save` to write the parameters to the POP backend and retrieve the draft order number `billCode`. This step is equivalent to the "Save" button in the POP backend interface, used to reserve the order number for secondary backend verification.

[0104] S2.9, Formal Submission: The agent instance, with action=submit and carrying the billCode returned by S2.8, calls pop_delisting_tool again to formally submit the draft to the approval process.

[0105] S2.10, Result Return and Exception Handling: Upon successful submission, the agent instance assembles the process number, current approval node, and estimated completion time returned by the POP backend into a message card and writes it back to the user's WeChat Work account. Upon submission failure, the agent instance automatically analyzes the error reason: if it is a parameter / data issue (such as EAN having no inventory record in the warehouse or an effective date earlier than the current date), it guides the user to supplement and modify the information and then returns to S2.4 for retry; if it is a system / network issue, it executes S2.9 again; if the retry limit is exceeded or it is not a data issue, it returns a failure message on the WeChat Work side and terminates the process.

[0106] Through the above process, this application achieves the following technical effects in the scenario of removing POP products from shelves: (1) Even if an attacker forges a WeChat message or a smart agent instance call chain, they cannot use other people's credentials to call pop_delisting_tool without authorization; the dual verification of PFCG role and MCP whitelist further ensures field-level permission isolation.

[0107] (2) By using the skill-based strong constraint clause, "manual confirmation" is made as a prerequisite for calling any write operation MCP tool. The "order" and "delivery" are explicitly separated by the multi-step arrangement between saving and submitting, which avoids the illusion of a large model directly triggering irreversible business operations. If any step fails, the specific node can be located and automatically recovered or the user can be guided to repair.

[0108] (3) Operational efficiency has been greatly improved. The original steps of “login to the backend, query EAN, query the warehouse, query the dictionary, enter more than ten items, temporarily save, and submit”, which took a total of 3 to 5 minutes and more than ten clicks, have been simplified to a single natural language dialogue. The average time has been shortened to less than 30 seconds. Furthermore, the probability of incorrect or missing fields has been significantly reduced through rule verification and auxiliary tools.

[0109] Please see Figure 8 This application also provides a multi-agent-based SAP system call and data interaction device, the device comprising: The agent determination module is used to obtain natural language information sent by the target object and a pre-configured routing mapping table, determine the target agent instance based on the natural language information, and search for the permission record of the target object in the routing mapping table; The permission confirmation and request generation module is used to input the natural language information into the target intelligent agent instance and generate an interface function module if the target intelligent agent instance has a corresponding record in the permission record. The credential generation module is used to load the interface credential corresponding to the interface function module from the pre-configured connection pool; The calling module is used to send the interface credentials to the SAP system, and after the SAP system returns authentication success information, it sends the interface function module to the SAP system.

[0110] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0111] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0112] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0113] Please see Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 using the methods described in the embodiments of this application. The input / output interface 903 is used to implement information input and output; The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904); The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0114] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0115] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0116] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0117] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0118] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0119] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0120] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0121] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0122] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0123] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0124] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0125] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0126] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0127] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0128] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0129] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for SAP system invocation and data interaction based on multi-agent systems, characterized in that, The method includes: Obtain the natural language information sent by the target object and the pre-configured routing mapping table, determine the target intelligent agent instance based on the natural language information, and look up the permission record of the target object in the routing mapping table; If the target agent instance has a corresponding record in the permission record, the natural language information is input into the target agent instance to generate an interface function module; Load the interface credentials corresponding to the interface function module into the pre-configured connection pool; The interface credentials are sent to the SAP system, and after the SAP system returns authentication success information, the interface function module is sent to the SAP system.

2. The method as described in claim 1, characterized in that, The method further includes: Obtain the configuration parameters of several intelligent agent instances, and configure several connection pools corresponding to each of the several intelligent agent instances according to the configuration parameters; Before sending the interface credentials to the SAP system, a connection object is created or returned from the connection pool corresponding to the target agent instance; Establish a session with the SAP system based on the connection object, and update the transaction context of the session.

3. The method as described in claim 1, characterized in that, The step of inputting the natural language information into the target intelligent agent instance to generate the interface function module includes: The target agent instance is used to automatically scan the trigger scenario descriptions of each skill in the loaded skill set during the inference process; The target intelligent agent instance is used to determine whether the skill with the highest matching degree is greater than a preset matching degree threshold. If so, the skill with the highest matching degree is loaded, and the interface function module is generated based on the natural language information. If not, initiate a dialogue with the target object to obtain new natural language information.

4. The method as described in claim 3, characterized in that, The skill also includes a processing flow, which includes parsing rules, verification rules, and function names; The process of loading the skill with the highest matching degree and generating the interface function module based on the natural language information includes: The parsing rules are determined, and the natural language information is extracted according to the parsing rules through the target intelligent agent instance to obtain the input parameters; Determine the verification rules, and perform integrity and format verification on the input parameters according to the verification rules using a target intelligent agent instance; Upon successful verification, the function name is determined, and the interface function module is generated based on the function name in the processing flow and the input parameters.

5. The method as described in claim 4, characterized in that, The step of performing completeness and format checks on the input parameters using the target agent instance according to the verification rules includes: The missing parameters are determined according to the verification rules. The next round of dialogue is generated based on the missing parameters. The natural language information is updated. The execution steps are returned: the parsing rules are determined, and the natural language information is extracted by the target intelligent agent instance according to the parsing rules to obtain the input parameters.

6. The method as described in claim 4, characterized in that, The method further includes: When the business scenario is updated, the dictionary specification corresponding to the transaction code is obtained, the processing flow and the triggering scenario description of the new skill are determined according to the dictionary specification, and the new skill is generated through the processing flow and the triggering scenario description.

7. The method as described in claim 4, characterized in that, The method further includes: When the interface definition of the SAP system is updated, the new function name is obtained, and the old function name in the skill is replaced by the new function name.

8. A multi-agent-based SAP system call and data interaction device, characterized in that, The device includes: The agent determination module is used to obtain natural language information sent by the target object and a pre-configured routing mapping table, determine the target agent instance based on the natural language information, and search for the permission record of the target object in the routing mapping table; The permission confirmation and request generation module is used to input the natural language information into the target intelligent agent instance and generate an interface function module if the target intelligent agent instance has a corresponding record in the permission record. The credential generation module is used to load the interface credential corresponding to the interface function module from the pre-configured connection pool; The calling module is used to send the interface credentials to the SAP system, and after the SAP system returns authentication success information, it sends the interface function module to the SAP system.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing 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 7.

Images