Operation and maintenance method and device based on large language model and computer program product

Abstract

The invention provides an operation and maintenance method and device based on a large language model, electronic equipment, a storage medium and a computer program product, relates to the technical field of computers, in particular to the technical field of artificial intelligence large models, natural language understanding and operation and maintenance, and can be applied to operation and maintenance scenes. The specific implementation scheme is as follows: iteratively executing the following operations through an intelligent agent: determining a link operation and maintenance result corresponding to a completed link in an operation and maintenance task; through a large language model, according to the link operation and maintenance result corresponding to the completed link, determining a target tool for continuing to execute the operation and maintenance task from the tool set; and calling the target tool, and generating a link operation and maintenance result corresponding to the next link in the operation and maintenance task. According to the method and the device, the Agent Loop (agent loop) mode is adopted, the target tool of the next link in the operation and maintenance task is determined according to the link operation and maintenance result corresponding to the completed link in the operation and maintenance task, and the applicability of the operation and maintenance scheme in each operation and maintenance scene and the accuracy of the operation and maintenance result are improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technologies, specifically to the fields of large artificial intelligence models, natural language understanding, and operation and maintenance technologies. In particular, it relates to an operation and maintenance method, device, electronic device, storage medium, and computer program product based on a large language model, which can be applied to operation and maintenance scenarios. Background Art

[0002] Network operation and maintenance mainly refers to the daily maintenance, management, monitoring, and optimization of networks, etc., to ensure the stable, efficient, and secure operation of the networks, and provide reliable support for various network applications and services. It covers aspects such as the configuration, debugging, and management of network devices (such as routers, switches, servers, etc.), the monitoring and analysis of network performance, the troubleshooting and resolution of faults, and the formulation and implementation of network security policies. Current network operation and maintenance solutions rely on a predefined rule library for alarm correlation to handle simple operation and maintenance events. Summary of the Invention

[0003] The present disclosure provides an operation and maintenance method, device, electronic device, storage medium, and computer program product based on a large language model.

[0004] According to a first aspect, there is provided an operation and maintenance method based on a large language model, including: iteratively performing the following operations by an agent: determining the operation and maintenance result corresponding to the completed link in the operation and maintenance task; determining, by the large language model, a target tool for continuing to execute the operation and maintenance task from a tool set according to the operation and maintenance result corresponding to the completed link; and invoking the target tool to generate the operation and maintenance result corresponding to the next link in the operation and maintenance task.

[0005] According to a second aspect, there is provided an operation and maintenance device based on a large language model, including the following processing units arranged in the agent to perform iterative operations: a result determination unit configured to determine the operation and maintenance result corresponding to the completed link in the operation and maintenance task; a tool determination unit configured to determine, by the large language model, a target tool for continuing to execute the operation and maintenance task from a tool set according to the operation and maintenance result corresponding to the completed link; and a result generation unit configured to invoke the target tool to determine the operation and maintenance result corresponding to the next link in the operation and maintenance task.

[0006] According to a third aspect, there is provided an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method described in any implementation manner of the first aspect.

[0007] According to a fourth aspect, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method described in any implementation manner of the first aspect.

[0008] According to a fifth aspect, there is provided a computer program product, including: a computer program which, when executed by a processor, implements the method described in any implementation manner of the first aspect.

[0009] According to the technology of the present disclosure, there is provided an operation and maintenance method and apparatus based on a large language model. By adopting the AgentLoop (intelligent agent loop) method, according to the link operation and maintenance results corresponding to the completed links in the operation and maintenance task, the target tool for the next link in the operation and maintenance task is determined, so as to call the target tool to complete the next link in the operation and maintenance task, improving the applicability of the operation and maintenance solution in various operation and maintenance scenarios and the accuracy of the operation and maintenance results.

[0010] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The drawings are used to better understand the solution and do not constitute a limitation to the present disclosure. Among them: Figure 1 is an exemplary system architecture diagram to which an embodiment of the present disclosure can be applied; Figure 2 is a flowchart of an embodiment of the operation and maintenance method based on a large language model according to the present disclosure; Figure 3 is a schematic diagram of an application scenario of the operation and maintenance method based on a large language model according to this embodiment; Figure 4 is a flowchart of another embodiment of the operation and maintenance method based on a large language model according to the present disclosure; Figure 5 is a structural diagram of an embodiment of the operation and maintenance apparatus based on a large language model according to the present disclosure; Figure 6 is a schematic structural diagram of a computer system suitable for implementing the embodiments of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0012] The following describes exemplary embodiments of the present disclosure with reference to the accompanying drawings. Various details of the embodiments of the present disclosure are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, descriptions of well-known functions and structures are omitted below for clarity and conciseness.

[0013] In the technical solution of the present disclosure, the processing of the collection, storage, use, processing, transmission, provision, and disclosure of the user's personal information complies with the provisions of relevant laws and regulations and does not violate public order and good customs.

[0014] Figure 1 An exemplary architecture 100 is shown to which the operation and maintenance method and device based on a large language model of the present disclosure can be applied.

[0015] As Figure 1 shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The terminal devices 101, 102, 103 are communicatively connected to form a topology network, and the network 104 is used as a medium to provide a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0016] The terminal devices 101, 102, 103 may be hardware devices or software that support network connections for data interaction and data processing. When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices that support network connections, information acquisition, interaction, display, processing, etc., including but not limited to smart phones, tablet computers, e-book readers, laptop computers, and desktop computers, etc. When the terminal devices 101, 102, 103 are software, they may be installed in the above-listed electronic devices. It may be implemented as, for example, multiple software or software modules for providing distributed services, or may be implemented as a single software or software module. No specific limitation is made here.

[0017] The server 105 may be a server that provides various services. For example, it is a background processing server that obtains the operation and maintenance tasks sent by the terminal devices 101, 102, 103 and determines the corresponding operation and maintenance results of the links in the operation and maintenance tasks through a large language model. Optionally, the server may feedback the operation and maintenance results of the links to the terminal devices. As an example, the server 105 may be a cloud server.

[0018] It should be noted that the server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster composed of multiple servers or as a single server. When the server is software, it can be implemented as multiple software or software modules (such as software or software modules for providing distributed services) or as a single software or software module. Specific limitations are not made here.

[0019] It should also be noted that the operation and maintenance method based on the large language model provided by the embodiments of the present disclosure is generally executed by the server, but the possibility of being executed by the terminal device or by the cooperation of the server and the terminal device is not excluded. Correspondingly, each part (such as each unit) included in the operation and maintenance device based on the large language model can be all set in the server, can be all set in the terminal device, or can be respectively set in the server and the terminal device.

[0020] It should be understood that Figure 1 the numbers of the terminal devices, networks, and servers in

[0021] are merely illustrative. According to the implementation requirements, there can be any number of terminal devices, networks, and servers. When the electronic device on which the operation and maintenance method based on the large language model runs does not need to perform data transmission with other electronic devices, the system architecture can only include the electronic device (such as a terminal device or a server) on which the operation and maintenance method based on the large language model runs. Figure 2 Figure 2 Please refer to which is a flowchart of an operation and maintenance method based on the large language model provided by the embodiments of the present disclosure. The agent iteratively executes process 200. Among them, process 200 includes the following steps:

[0022] In this embodiment, the execution subject of the operation and maintenance method based on the large language model (for example, Figure 1 the server in

[0023] ​Operation and maintenance tasks include, but are not limited to, network operation and maintenance tasks, system operation and maintenance tasks, application operation and maintenance tasks, data operation and maintenance tasks, security operation and maintenance tasks, and cloud operation and maintenance tasks. Network operation and maintenance tasks, for example, include the management and maintenance of network devices, network performance monitoring and optimization, network fault troubleshooting and handling, network security maintenance, network optimization and upgrade, etc.; system operation and maintenance tasks, for example, include the management and maintenance of servers, operating systems, and file systems; application operation and maintenance tasks, for example, include the deployment, monitoring, and maintenance of application programs, middleware management and maintenance; data operation and maintenance tasks, for example, include database management and maintenance, data warehouse and data lake management, data quality management; security operation and maintenance tasks, for example, include security monitoring and auditing, vulnerability management, security policy formulation and implementation; cloud operation and maintenance tasks, for example, include cloud resource management, cloud service management.

[0024] A complete operation and maintenance task generally includes multiple links. Taking the network vulnerability scanning and repair task as an example, it includes the scanning and evaluation link, the repair plan formulation link, and the repair implementation and verification link; taking the network device inspection task, it includes the device status inspection link, the performance index monitoring link, and the configuration verification link.

[0025] In this embodiment, an intelligent agent is set in the above execution subject. In the process of determining the link operation and maintenance result corresponding to each link in the operation and maintenance task, the intelligent agent executes steps 201-203, that is, the intelligent agent first executes the above step 201 to determine the link operation and maintenance result corresponding to the completed link in the operation and maintenance task; then, inputs the link operation and maintenance result corresponding to the completed link into the large language model. Based on the subsequent step 202, the large language model determines the target tool for the next link in the operation and maintenance task according to the link operation and maintenance result corresponding to the completed link in the operation and maintenance task; finally, the intelligent agent calls the target tool to generate the link operation and maintenance result corresponding to the next link in the operation and maintenance task. Among them, the completed link represents the link whose corresponding link operation and maintenance result is generated by the target tool determined by the large language model. The next link is the next uncompleted link after the completed link. In this way, a part of the operation and maintenance task corresponding to one link in the operation and maintenance task is completed.

[0026] By iteratively executing steps 201 - 203, the links in the operation and maintenance tasks are sequentially processed by the agent. The above - mentioned processing method of the agent's iteration is the Agent Loop. As an example, the above - mentioned execution entity receives the model prompt data (such as prompt words) of the large - language model. Based on the powerful natural - language understanding ability and data - analysis ability of the large - language model, it determines the operation and maintenance tasks represented by the model prompt data; splits the operation and maintenance tasks to obtain multiple links in series, in parallel, or a combination of series and parallel; then, according to the logical relationship between the multiple links, determines the processing order of the links; and processes each link in sequence according to the processing order. For the first link, there is no corresponding completed link, that is, the link operation and maintenance result of the completed link is empty. The large - language model can directly determine the target tool according to the model prompt data representing the operation and maintenance tasks and the relevant information of the first link, so as to obtain the link operation and maintenance result of the first link through the subsequent steps 202 and 203.

[0027] For non - first links in the operation and maintenance tasks, in response to the execution of the link, determine the link operation and maintenance results corresponding to the completed links in the operation and maintenance tasks.

[0028] In some optional implementation manners of this embodiment, the above - mentioned execution entity can execute the above - mentioned step 201 in the following manner: Determine the link operation and maintenance results corresponding to the preset number of completed links up to the current in the operation and maintenance tasks.

[0029] The preset number can be flexibly set according to the actual situation. For example, the preset number is 10.

[0030] By setting the preset number, the number of link operation and maintenance results input to the large - language model can be restricted. In some cases, there are many links included in the operation and maintenance tasks. As the operation and maintenance tasks are executed, the link operation and maintenance results corresponding to the completed links are more and more. And the link operation and maintenance results corresponding to the completed links sorted in the front may not help the data - processing process of the large - language model, but instead impose a large data - processing pressure on the large - language model.

[0031] In this implementation manner, restricting the number of link operation and maintenance results input to the large - language model by the preset number reduces the data - processing pressure of the large - language model while ensuring the accuracy of the processing results of the large - language model, and improves the data - processing efficiency of the large - language model.

[0032] Step 202, through the large - language model, determine the target tool for continuing to execute the operation and maintenance tasks from the tool set according to the link operation and maintenance results corresponding to the completed links.

[0033] In this embodiment, the above - mentioned execution entity can determine the target tool for continuing to execute the operation and maintenance tasks from the tool set through the large - language model according to the link operation and maintenance results corresponding to the completed links.

[0034] The tool set includes tools for performing operation and maintenance tasks. You can set up corresponding tool sets for different types of operation and maintenance tasks. Taking network operation and maintenance tasks as an example, the tool set includes equipment connection and login tools, equipment status monitoring tools, equipment log analysis tools, etc. for network equipment inspection, network traffic monitoring tools, performance indicator monitoring tools, etc. for network performance monitoring, network connectivity testing tools, network packet capture analysis tools for network troubleshooting, vulnerability scanning tools, intrusion detection and defense tools, etc. for network security maintenance.

[0035] For example, the division granularity (precision) of the types of operation and maintenance tasks can be specifically set according to the actual situation, so as to set a tool set in a targeted manner corresponding to the type of operation and maintenance tasks. Specifically, the operation and maintenance tasks can be divided into multiple nested levels of different granularities based on the preset division levels, and the correspondence between the types of operation and maintenance tasks and tools at the minimum level can be determined; then, the user can select the required type of operation and maintenance tasks according to their own needs, thereby obtaining a tool set consisting of tools corresponding to the selected type of operation and maintenance tasks. The tools corresponding to the type of operation and maintenance tasks are used to process the type of operation and maintenance tasks.

[0036] For another example, the tool set includes tools applicable to various operation and maintenance tasks, so that the tool set has universal applicability to various operation and maintenance tasks.

[0037] The tools in the tool collection can be tools that support the MCP (Model Communication Protocol) protocol, or tools customized by users according to their own needs. The MCP protocol is a protocol for communication between models. It defines the rules for how models interact, share data, and work together so that different models can seamlessly exchange information and collaborate.

[0038] The tools in the tool collection can be implemented in the process of the agent, or integrated with the agent as external services through HTTP (Hypertext Transfer Protocol), RPC (Remote Procedure Call), etc. Implementing tools in the process of the agent means developing and integrating tools in the same process as the agent. This method has the advantages of low latency and efficient calling, because the tools and the agent share the same memory space and system resources, and data exchange does not require serialization and network transmission.

[0039] A tool as an external service refers to a tool corresponding to an external service called through protocols such as HTTP and RPC. In this way, the tool is deployed as an independent service, and the agent interacts with it through standard communication protocols. Its advantages include decoupling, high flexibility, easy expansion and maintenance, and different services can be developed, deployed, and expanded independently.

[0040] In this embodiment, the link operation and maintenance result corresponding to the completed link is input into the large language model. Based on its powerful natural language understanding ability and data analysis ability, the large language model determines the target tool for continuing to execute the operation and maintenance task from the tool set.

[0041] The links in an operation and maintenance task may change. Continuing with the example of the network vulnerability scanning and repair task, it generally includes a scanning and assessment link, a repair plan formulation link, and a repair implementation and verification link. However, when the link operation result corresponding to the scanning and assessment link indicates that no vulnerability is found, the subsequent repair plan formulation link and repair implementation and verification link will no longer be executed.

[0042] For an operation and maintenance task whose included links may change, the above-mentioned execution entity can, through the large language model, determine the next link to be executed based on the link operation and maintenance result corresponding to the completed link; and based on the determined next link and the link operation and maintenance result corresponding to the completed link, determine the target tool for continuing to execute the operation and maintenance task from the tool set.

[0043] For an operation and maintenance task whose included links are fixed, the above-mentioned execution entity can directly, through the large language model, determine the target tool for continuing to execute the operation and maintenance task from the tool set based on the link operation and maintenance result corresponding to the completed link and the next link to be executed determined based on the splitting operation.

[0044] In some optional implementation manners of this embodiment, the above-mentioned execution entity can also execute step 202 in the following manner: through the large language model, determine the target tool from the tool set according to the model prompt data representing the operation and maintenance task and the link operation and maintenance result corresponding to the completed link.

[0045] Regarding the links in the operation and maintenance task, when determining the target tool corresponding to a link, it is necessary to input the model prompt data representing the operation and maintenance task and the link operation and maintenance result corresponding to the completed link into the large language model. The large language model combines the model prompt data and the link operation and maintenance result corresponding to the completed link to determine the target tool for continuing to execute the operation and maintenance task from the tool set.

[0046] When the steps in the operation and maintenance task may change, the large language model determines the next step to be executed based on the model prompt data and the step operation and maintenance results corresponding to the completed steps; based on the prompt data, the determined next step, and the step operation and maintenance results corresponding to the completed steps, it determines the target tool for continuing to execute the operation and maintenance task from the tool set.

[0047] For operation and maintenance tasks with fixed steps, the above-mentioned execution entity can directly use the large language model to determine the target tool for continuing to execute the operation and maintenance task from the tool set based on the model prompt data, the step operation and maintenance results corresponding to the completed steps, and the next step to be executed.

[0048] In this implementation, on the basis of considering the step operation and maintenance results corresponding to the completed steps, the model prompt data is further considered. Under the guidance of the model prompt data, the accuracy of determining the target tool can be further improved.

[0049] In some optional implementations of this embodiment, the above-mentioned execution entity can execute the process of determining the target tool in the following way: using the large language model, based on the model prompt data, the step operation and maintenance results corresponding to the completed steps, and the configuration data of the tools in the tool set, it determines the target tool from the tool set.

[0050] The configuration data of the tool is a set of various parameters and settings required for the tool to run normally and execute tasks, which helps the large language model understand the functions of the tool and be able to accurately call the tool.

[0051] The above-mentioned intelligent agent inputs the model prompt data, the step operation and maintenance results corresponding to the completed steps, and the configuration data of the tools in the tool set into the large language model. The large language model deeply understands and analyzes these three types of data and determines the target tool from the tool set.

[0052] In this implementation, the large language model further considers the configuration data of the tools on the basis of considering the model prompt data and the step operation and maintenance results corresponding to the completed steps, enabling the large language model to determine the functions of each tool according to the configuration data of the tools in the tool set to determine the tool suitable for the next step, further improving the accuracy of the target tool.

[0053] In some alternative implementation manners of this embodiment, the configuration data includes tool description data and tool parameter data. The tool description data is used to explain the main purpose of the tool and illustrate the specific problems that the tool can solve. In some more detailed tool description data, it also includes data such as the basic operation mechanism of the tool and the operation steps of the tool. The tool parameter data refers to the configuration options and input values required for the tool to run or execute a specific task. These parameters are used to customize the behavior of the tool so that it can work according to the specific needs and environment of the user.

[0054] In this implementation manner, the above-mentioned execution subject can execute the determination process of the target tool in the following way: through the large language model, based on the model prompt data, the link operation and maintenance results corresponding to the completed links, and the tool description data and tool parameter data of the tools in the tool set, determine the target tool from the tool set, and determine the target parameter value corresponding to the tool parameter data of the target tool.

[0055] The above-mentioned intelligent agent inputs the model prompt data, the link operation and maintenance results corresponding to the completed links, and the tool description data and tool parameter data of the tools in the tool set into the large language model. The large language model determines the target tool from the tool set and determines the target parameter value corresponding to the tool parameter data based on the tool parameter data of the target tool, so that the target tool can execute the next link of the operation and maintenance task based on the target parameter value.

[0056] As an example, first, the large language model identifies information such as the name, type, and value range of each parameter according to the tool parameter data of the target tool. Then, match the link operation and maintenance results of the completed links with the parameter requirements of the target tool to find appropriate initial values for each parameter. Finally, adjust and optimize the parameter values according to the actual situation and historical experience to ensure that the tool can better complete the task.

[0057] In this implementation manner, based on determining the target tool for the next link, the large language model further determines the target parameter value corresponding to the tool parameter data of the target tool, improving the richness of the generated information and providing accurate data basis for the subsequent tool call process.

[0058] In some alternative implementation manners of this embodiment, the above-mentioned execution subject can also perform the following operations: set the configuration data of the tools in the tool set through at least one of the model context protocol, configuration file, and database.

[0059] MCP is a protocol used to transfer information in the model context. The configuration data of the tool can be transferred from the relevant service or module to the large language model that needs to use the tool through MCP. When performing tasks, the large language model can utilize the configuration data received by MCP to select appropriate tools to assist in task completion.

[0060] The configuration method based on the model context protocol can update and transfer tool information to the large language model in real time, and can dynamically adjust the transferred configuration data according to the requirements of the large language model or the scenarios of operation and maintenance tasks.

[0061] In the configuration method based on the configuration file, the configuration data of the tool is stored in the configuration file in the JSON (JavaScript Object Notation) format. JSON is a lightweight data exchange format, which is easy to read and write, and is also easy for machines to parse and generate. Information such as the name, description, and parameters of the tool is organized in key-value pairs in the JSON file.

[0062] In the configuration method based on the database, a database is used to store the configuration data of the tool. A database table structure can be designed to save data such as the name of the tool, tool description data, and tool parameter data. For example, a configuration data table can have the following fields: tool ID (Identity Document), tool name, tool description, parameter definition (which can be a JSON string field to store detailed information about the parameters, such as parameter name, type, description, etc.). The list of tool information is obtained through database query statements and then organized into JSON format to be provided to the model or other components that need to use tool information.

[0063] In this implementation method, multiple configuration file configuration methods are provided, which improves the flexibility of the configuration process.

[0064] Step 203: Invoke the target tool to generate the corresponding link operation and maintenance result for the next link in the operation and maintenance task.

[0065] In this embodiment, the above execution entity can invoke the target tool to generate the corresponding link operation and maintenance result for the next link in the operation and maintenance task.

[0066] Continuing with the network vulnerability scanning and repair task as an example, which includes the scanning and assessment phase, the repair plan formulation phase, and the repair implementation and verification phase. In the scanning and assessment configuration phase, the corresponding target tool is Nessus (a vulnerability scanning tool). Nessus sends various detection requests to the target device corresponding to the operation and maintenance task, simulating potential attack behaviors to identify vulnerabilities on the device. During the scanning process, Nessus will display information such as the scanning progress and the number of discovered vulnerabilities in real time. After the scanning is completed, Nessus will generate a detailed scanning report, including information such as the name, description, severity, and affected devices of the vulnerabilities. According to the classification of the vulnerability severity in the report, focus on high-risk vulnerabilities, and analyze the potential impact of each vulnerability, such as whether it may cause system crashes, data leaks, etc.

[0067] In the repair plan formulation phase, the corresponding target tool is Qualys Vulnerability Management. First, the Qualys vulnerability management tool will further analyze and classify the imported vulnerability data. According to factors such as the severity of the vulnerability, the difficulty of exploitation, and the importance of the affected assets, the vulnerabilities are prioritized. For example, high-risk vulnerabilities that can be easily exploited and affect critical business systems will be given higher priorities. Then, based on the analysis results, Qualys generates repair suggestions for each vulnerability. These repair suggestions include specific repair steps, official patch links, configuration modification methods, etc. At the same time, considering the technical environments and business requirements of different enterprises, the repair suggestions may provide multiple optional solutions, such as installing patches immediately, making temporary configuration changes, etc. Finally, combining the actual situation of the enterprise and the repair suggestions, use Qualys to formulate a detailed repair plan. Determine the repair time and responsible person for each vulnerability, reasonably arrange the repair order to ensure that high-risk vulnerabilities can be repaired in a timely manner. At the same time, set corresponding reminder and tracking mechanisms for each repair task to ensure the smooth progress of the repair work.

[0068] In the repair implementation and verification phase, the corresponding target tool is MS Patch Management. First, obtain the corresponding vulnerability patches according to the patch links or information provided in the repair plan. Before officially deploying the patches, establish a test environment similar to the production environment, install the patches on the devices in the test environment, and conduct comprehensive functional and stability tests. The test content includes checking whether the system runs normally, whether the applications are compatible, and whether the business processes are affected. Then, after the tests pass, use the MS Patch Management tool to deploy the patches to the target devices in the production environment. You can choose methods such as batch deployment or gradual deployment according to the predetermined deployment plan and strategy to ensure the smooth progress of the patch installation process. At the same time, closely monitor the status of the devices during the patch installation process and promptly handle possible installation failures or system anomalies. Then, after the patch installation is completed, use a vulnerability scanning tool (such as Nessus) to scan the repaired devices again to verify whether the vulnerabilities have been successfully repaired. Check the scan report to confirm that the vulnerability status has changed from "unrepaired" to "repaired". At the same time, manually check the configuration and functions of the devices to ensure the normal operation and security of the system. Finally, record the entire repair process, including the repair time, devices, vulnerability information, patch version, etc. If problems occur during the repair process or the patches cause new failures, promptly initiate a rollback plan to restore the system to its previous stable state. You can use the previously backed-up system configuration or data for the rollback operation to ensure business continuity and stability.

[0069] In this embodiment, the above-mentioned execution subject may iteratively execute the above steps 201-203 until a preset end condition is reached. The preset end condition includes but is not limited to completing the operation and maintenance tasks, the number of tool invocations exceeding a preset threshold, and determining that there is no target tool corresponding to the next link in the tool set according to the link operation and maintenance results corresponding to the completed links.

[0070] In some optional implementation manners of this embodiment, the above-mentioned execution subject may execute the above step 203 in the following manner: Invoke the target tool with the target parameter value to generate the link operation and maintenance result corresponding to the next link in the operation and maintenance task.

[0071] In this implementation manner, the target tool with the target parameter value is more suitable for the next link in the operation and maintenance task, which helps to further improve the accuracy of the link operation and maintenance result corresponding to the next link.

[0072] In some optional implementation manners of this embodiment, the above-mentioned execution subject may execute the generation process of the link operation and maintenance result in the following manner: The first step is to determine the target data source according to the target tool and the target parameter value.

[0073] As an example, first, analyze the configuration options of the target tool to determine the types of data sources it supports (such as local files, databases, web services, etc.). Then, locate the specific data source location based on the tool parameter values and the data analysis requirements of the target tool; finally, check the availability of the data source and the integrity of the data to ensure that the data can be correctly read.

[0074] As another example, first, clarify the sources of data required for each link according to the business logic and processes of the operation and maintenance tasks. Then, infer a reasonable data source based on the parameter values of the target tool and the business requirements. Finally, verify whether the inferred data source is reasonable and make adjustments if there are any deviations.

[0075] In the second step, call the target tool with the target parameter values, and generate the corresponding link operation and maintenance results for the next link in the operation and maintenance task based on the data in the target data source.

[0076] After determining the target data source, the above intelligent agent calls the target tool, which uses the target parameter values and generates the corresponding link operation and maintenance results for the next link in the operation and maintenance task based on the data in the target data source.

[0077] In this implementation, the most suitable analysis tools and data sources can be dynamically and intelligently selected and orchestrated. This "thinking-style" automated process makes the operation and maintenance solutions applicable to a wider and deeper range of problem scenarios.

[0078] In some optional implementations of this embodiment, the above execution entity can execute the second step in the following manner: during the processing of the data in the target data source by the target tool using the target parameter values, combine the large language model for auxiliary analysis and reasoning to generate the corresponding link operation and maintenance results for the next link in the operation and maintenance task.

[0079] During the information processing of the target tool, some or all of the data processing operations can be analyzed and inferred using the large language model. That is, the tool itself can call the large language model for reasoning and analysis. For example, the target tool can call the large language model through an API (Application Programming Interface); another example is that the SDK (Software Development Kit) of the large language model is embedded in the tool's code library, and inside the target tool, the data to be analyzed is sorted according to the format specified by the SDK. At the same time, the large language model is initialized and configured through the SDK, including setting model parameters (such as the maximum generation length, temperature value, etc.). The data is input into the embedded model, and the inference interface of the SDK is called to start the model analysis.

[0080] In this implementation manner, the data processing process of the target tool can be further assisted by a large language model for analysis and reasoning, which helps to further improve the accuracy of the link operation and maintenance results.

[0081] Continue to refer to Figure 3 , Figure 3 FIG. 300 is a schematic diagram of an application scenario of the operation and maintenance method based on a large language model according to this embodiment. The user 301 sends model prompt data representing an operation and maintenance task to the agent in the server 303 through the terminal device 302. The agent inputs the model prompt data into the large language model, and the large language model determines the target tool for the first link in the operation and maintenance task, that is, the target tool for the first link, and calls the target tool for the first link to generate the link operation and maintenance result corresponding to the first link in the operation and maintenance task, that is, the operation and maintenance result for the first link. In the subsequent processing process, the agent iteratively performs the following operations: determining the link operation and maintenance result corresponding to the completed link in the operation and maintenance task; through the large language model, determining the target tool for continuing to perform the operation and maintenance task from the tool set according to the link operation and maintenance result corresponding to the completed link; calling the target tool to generate the link operation and maintenance result corresponding to the next link in the operation and maintenance task.

[0082] For example, after completing the first link of the operation and maintenance task, the agent first determines the operation and maintenance result for the first link corresponding to the first link in the operation and maintenance task; then, through the large language model, determines the target tool for the second link for continuing to perform the operation and maintenance task from the tool set according to the operation and maintenance result for the first link corresponding to the first link; finally, calls the target tool for the second link to generate the operation and maintenance result for the second link corresponding to the second link in the operation and maintenance task.

[0083] After completing the second link of the operation and maintenance task, the agent first determines the operation and maintenance result for the first link corresponding to the first link in the operation and maintenance task and the operation and maintenance result for the second link corresponding to the second link; then, through the large language model, determines the target tool for the third link for continuing to perform the operation and maintenance task from the tool set according to the operation and maintenance result for the first link corresponding to the first link and the operation and maintenance result for the second link corresponding to the second link; finally, calls the target tool for the third link to generate the operation and maintenance result for the third link corresponding to the third link in the operation and maintenance task.

[0084] In this embodiment, an operation and maintenance method based on a large language model is provided. By adopting the Agent Loop (agent loop) method, according to the link operation and maintenance result corresponding to the completed link in the operation and maintenance task, the target tool for the next link in the operation and maintenance task is determined, so as to call the target tool to complete the next link in the operation and maintenance task, improving the applicability of the operation and maintenance plan in various operation and maintenance scenarios and the accuracy of the operation and maintenance result.

[0085] In some alternative implementation manners of this embodiment, the above-mentioned execution entity may further perform the following operations: During the execution of the operation and maintenance task, in response to the number of tool invocations up to the current time exceeding a preset number threshold, generate an overall operation and maintenance result according to the operation and maintenance result corresponding to the completed link.

[0086] The preset number threshold can be flexibly set according to the actual situation. For example, the number of tool invocations is 20.

[0087] By setting the preset number threshold, the number of times the large language model invokes tools can be restricted, avoiding the hallucination of the large language model and always returning a callable tool. During the execution of the operation and maintenance task, when the number of tool invocations reaches the preset number threshold, the agent no longer starts a new invocation but exits the agent loop and returns a prompt message for exiting the agent loop to the user.

[0088] The large language model can summarize and analyze the operation and maintenance results corresponding to the completed links to obtain the overall operation and maintenance result.

[0089] In this implementation manner, by setting the preset number threshold, the number of times the large language model invokes tools is restricted, avoiding the hallucination problem of the large language model and ensuring the effectiveness of the data processing process of the large language model.

[0090] In some alternative implementation manners of this embodiment, the above-mentioned execution entity may further perform the following operations: In response to determining, through the large language model, that there is no target tool in the tool set according to the operation and maintenance result corresponding to the completed link, generate an overall operation and maintenance result according to the operation and maintenance result corresponding to the completed link.

[0091] When the large language model determines that there is no target tool in the tool set according to the operation and maintenance result corresponding to the completed link, it indicates that the operation and maintenance task cannot be continued. The large language model can summarize and analyze the operation and maintenance result corresponding to the completed link to obtain the overall operation and maintenance result.

[0092] In this implementation manner, a method for generating the overall operation and maintenance result is provided, improving the comprehensiveness of the output processing.

[0093] In some alternative implementation manners of this embodiment, the above-mentioned execution entity may further perform the following operations: In response to completing the operation and maintenance task, determine the overall operation and maintenance result of the operation and maintenance task according to the operation and maintenance result corresponding to the last link of the operation and maintenance task.

[0094] Complete the characterization of the operation and maintenance tasks. Through the tools corresponding to each link in the operation and maintenance tasks, part of the operation and maintenance tasks of the complete link are obtained to get the link processing results. In this case, the above-mentioned execution entity can directly determine the link operation and maintenance result corresponding to the last link of the operation and maintenance task as the total operation and maintenance result of the operation and maintenance task. Or, based on emphasizing the link operation and maintenance result corresponding to the last link, the large language model can summarize and analyze the data in combination with the link operation and maintenance results corresponding to other completed links to obtain the total operation and maintenance result.

[0095] In this implementation method, a method for generating the total operation and maintenance result is provided, which improves the comprehensiveness of the output processing.

[0096] In some optional implementation methods of this embodiment, the above-mentioned execution entity can also perform the following operations: call the first output tool to generate and display the total result data in the first specified format according to the total operation and maintenance result of the operation and maintenance task; and / or call the second output tool to generate and display the link result data in the second specified format according to the link operation and maintenance result.

[0097] Among them, both the first output tool and the second output tool are tools with the function of outputting data results. The first output tool and the second output tool can be the same output tool or different output tools. For different links in the operation and maintenance tasks, the corresponding second output tools can be the same or different. In this implementation method, the destination node of the output data can also be determined according to user needs, including but not limited to IM (Instant Messaging) work groups, text messages, phone calls, web pages, etc.

[0098] In this implementation method, the following at least one output method can be flexibly selected according to requirements: 1. After the operation and maintenance task is completed, call the first output tool to generate and display the total result data in the first specified format according to the total operation and maintenance result of the operation and maintenance task.

[0099] 2. During the execution of the operation and maintenance task, call the second output tool to generate and display the link result data in the second specified format according to the link operation and maintenance result.

[0100] 3. During the execution of the operation and maintenance task, call the second output tool to generate and display the link result data in the second specified format according to the link operation and maintenance result; after the operation and maintenance task is completed, call the first output tool to generate and display the total result data in the first specified format according to the total operation and maintenance result of the operation and maintenance task.

[0101] In this implementation method, the output methods of the link operation and maintenance result and the total result data can be flexibly set, which helps to improve the user's information acquisition efficiency and experience.

[0102] Continue to refer toFigure 4 , shows a schematic process 400 of another embodiment of the operation and maintenance method based on a large language model according to the present disclosure. In the process 400, the following steps are included: Step 401, perform the following operations through agent iteration: Step 4011, determining the link operation and maintenance results corresponding to the preset number of completed links in the operation and maintenance task up to the current time.

[0103] Step 4012, through the large language model, according to the model prompt data, the link operation and maintenance results corresponding to the completed link, and the tool description data and tool parameter data of the tools in the tool collection, determine the target tool from the tool collection, and determine the target parameter value corresponding to the tool parameter data of the target tool.

[0104] Step 4013, determine the target data source according to the target tool and the target parameter value.

[0105] Step 4014, calling the target tool using the target parameter value, and generating the link operation and maintenance result corresponding to the next link in the operation and maintenance task according to the data in the target data source.

[0106] Step 402, during the execution of the operation and maintenance task, in response to the number of tool calls to date exceeding a preset number threshold, a total operation and maintenance result is generated according to the operation and maintenance results of the links corresponding to the completed links.

[0107] Step 403, in response to determining through the large language model that the target tool does not exist in the tool set according to the link operation and maintenance results corresponding to the completed links, a total operation and maintenance result is generated according to the link operation and maintenance results corresponding to the completed links.

[0108] Step 404, in response to completing the operation and maintenance task, determining the overall operation and maintenance result of the operation and maintenance task according to the link operation and maintenance result corresponding to the last link of the operation and maintenance task.

[0109] The process 400 of the operation and maintenance method based on the large language model in this embodiment specifically illustrates the process of determining the link operation and maintenance results and the process of determining the overall operation and maintenance results. It adopts the Agent Loop method to determine the target tool for the next link in the operation and maintenance task based on the link operation and maintenance results corresponding to the completed links in the operation and maintenance task, so as to call the target tool to complete the next link in the operation and maintenance task, and generate the overall operation and maintenance results in a targeted manner based on different situations, thereby further improving the applicability of the operation and maintenance solution in various operation and maintenance scenarios and the accuracy of the operation and maintenance results.

[0110] To further illustrate the present disclosure, the operation and maintenance task is a latency analysis task in a high-performance network as an example to illustrate its specific data processing process: 1. The user queries the current latency data of the target training task related to the data center on the latency page; 2. Receive the user's click operation on the "Intelligent Analysis" button on the web page; 3. Transmit the ID and latency data of the target training task to the intelligent agent to start the intelligent agent loop; 4. The intelligent agent submits the prompt words and tool set representing latency analysis to the large language model; 5. The large language model returns the link operation result, and the link operation result includes calling the latency analysis tool and the parameter values of the latency analysis tool; 6. The latency analysis tool re-pulls the latency data of this target training task according to the ID and parameters of the target training task, and at the same time pulls the topology data of this target training task; 7. The latency analysis tool submits the latency data and topology data to the large language model, and the large language model generates an analysis result; 8. The large language model returns the latency analysis result of this target training task, and the latency analysis result is in the markdown (lightweight markup language) format; 9. The intelligent agent calls the web reporting tool to generate a professional and beautiful latency report from this markdown format result, upload the report to the specified URL (Uniform Resource Locator), and store it in the database; 10. The intelligent agent returns the URL of the report to the user.

[0111] Continue to refer to Figure 5 , as an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of an operation and maintenance device based on a large language model. This system embodiment corresponds to Figure 2 the method embodiment shown, and this system can be specifically applied to various electronic devices.

[0112] As Figure 5 shown, the operation and maintenance device 500 based on the large language model includes the following processing units that perform iterative operations in the intelligent agent: a result determination unit 501, configured to determine the link operation result corresponding to the completed link in the operation and maintenance task; a tool determination unit 502, configured to determine the target tool for continuing to execute the operation and maintenance task from the tool set through the large language model according to the link operation result corresponding to the completed link; a result generation unit 503, configured to call the target tool to determine the link operation result corresponding to the next link in the operation and maintenance task..

[0113] In some alternative implementation manners of this embodiment, the tool determination unit 502 is further configured to: determine a target tool from a tool set through a large language model according to model prompt data representing an operation and maintenance task and the link operation and maintenance result corresponding to the completed link.

[0114] In some alternative implementation manners of this embodiment, the tool determination unit 502 is further configured to: determine a target tool from a tool set through a large language model according to the model prompt data, the link operation and maintenance result corresponding to the completed link, and the configuration data of the tools in the tool set.

[0115] In some alternative implementation manners of this embodiment, the configuration data includes tool description data and tool parameter data, and the tool determination unit 502 is further configured to: determine a target tool from a tool set through a large language model according to the model prompt data, the link operation and maintenance result corresponding to the completed link, and the tool description data and tool parameter data of the tools in the tool set, and determine a target parameter value corresponding to the tool parameter data of the target tool.

[0116] In some alternative implementation manners of this embodiment, the result generation unit 503 is further configured to: call the target tool with the target parameter value to generate the link operation and maintenance result corresponding to the next link in the operation and maintenance task.

[0117] In some alternative implementation manners of this embodiment, the result generation unit 503 is further configured to: determine a target data source according to the target tool and the target parameter value; call the target tool with the target parameter value, and generate the link operation and maintenance result corresponding to the next link in the operation and maintenance task according to the data in the target data source.

[0118] In some alternative implementation manners of this embodiment, the result generation unit 503 is further configured to: during the processing of the data in the target data source by the target tool with the target parameter value, perform auxiliary analysis and reasoning in combination with the large language model to generate the link operation and maintenance result corresponding to the next link in the operation and maintenance task.

[0119] In some alternative implementation manners of this embodiment, the above device further includes: a tool configuration unit (not shown in the figure), configured to: set the configuration data of the tools in the tool set through at least one of a model context protocol, a configuration file, and a database.

[0120] In some alternative implementation manners of this embodiment, the result determination unit 501 is further configured to: determine the link operation and maintenance results corresponding to a preset number of completed links up to the current in the operation and maintenance task.

[0121] In some alternative implementation manners of this embodiment, the result generation unit 503 is further configured to: during the execution of the operation and maintenance task, in response to the number of tool calls up to the current time exceeding a preset number threshold, generate a total operation and maintenance result according to the operation and maintenance results corresponding to the completed links.

[0122] In some alternative implementation manners of this embodiment, the result generation unit 503 is further configured to: in response to determining, through a large language model, that there is no target tool in the tool set according to the operation and maintenance results corresponding to the completed links, generate a total operation and maintenance result according to the operation and maintenance results corresponding to the completed links.

[0123] In some alternative implementation manners of this embodiment, the result generation unit 503 is further configured to: in response to completing the operation and maintenance task, determine the total operation and maintenance result of the operation and maintenance task according to the operation and maintenance result corresponding to the last link of the operation and maintenance task.

[0124] In some alternative implementation manners of this embodiment, the above device further includes a result output unit (not shown in the figure), which is configured to: call a first output tool to generate and display total result data in a first specified format according to the total operation and maintenance result of the operation and maintenance task; and / or call a second output tool to generate and display link result data in a second specified format according to the operation and maintenance results of the links.

[0125] In this embodiment, an operation and maintenance device based on a large language model is provided. By adopting the Agent Loop (intelligent agent loop) method, according to the operation and maintenance results corresponding to the completed links in the operation and maintenance task, the target tool for the next link in the operation and maintenance task is determined, so as to call the target tool to complete the next link in the operation and maintenance task, improving the applicability of the operation and maintenance solution in various operation and maintenance scenarios and the accuracy of the operation and maintenance results.

[0126] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device, which includes: 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, and when the instructions are executed by the at least one processor, the at least one processor is enabled to implement the operation and maintenance method based on a large language model described in any of the above embodiments.

[0127] According to an embodiment of the present disclosure, the present disclosure further provides a readable storage medium, which stores computer instructions for enabling a computer to implement the operation and maintenance method based on a large language model described in any of the above embodiments when executed.

[0128] The embodiment of the present disclosure provides a computer program product, which can implement the operation and maintenance method based on a large language model described in any of the above embodiments when executed by a processor.

[0129] Figure 6 FIG. shows a schematic block diagram of an exemplary electronic device 600 that can be used to implement embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0130] As Figure 6 shown, the device 600 includes a computing unit 601 that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the device 600 can also be stored. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0131] A plurality of components in the device 600 are connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a magnetic disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0132] The computing unit 601 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 executes the various methods and processes described above, such as the operation and maintenance method based on the large language model. For example, in some embodiments, the operation and maintenance method based on the large language model can be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the operation and maintenance method based on the large language model described above can be executed. Alternatively, in other embodiments, the computing unit 601 can be configured to execute the operation and maintenance method based on the large language model in any other suitable manner (e.g., by means of firmware).

[0133] Various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a dedicated or general-purpose programmable processor, receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0134] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to the processor or controller of a general-purpose computer, a special-purpose computer, or other programmable operation and maintenance devices based on the large language model, such that when the program code is executed by the processor or controller, the functions / operations specified in the flowchart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, executed partially on the machine as an independent software package and partially on a remote machine, or executed entirely on a remote machine or server.

[0135] In the context of this disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0136] In order to provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0137] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0138] A computer system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The relationship between the client and the server is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability existing in traditional physical hosts and virtual private server (VPS) services; it may also be a server of a distributed system or a server combined with a blockchain.

[0139] According to the technical solution of the embodiment of the present disclosure, an operation and maintenance method and device based on a large language model are provided. By adopting the Agent Loop method, according to the link operation and maintenance result corresponding to the completed link in the operation and maintenance task, the target tool for the next link in the operation and maintenance task is determined, so as to call the target tool to complete the next link in the operation and maintenance task, improving the applicability of the operation and maintenance solution in various operation and maintenance scenarios and the accuracy of the operation and maintenance result.

[0140] It should be understood that various forms of the processes shown above can be used, reordering, adding or deleting steps. For example, the steps described in the present disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution provided by the present disclosure can be achieved, and no limitation is made herein.

[0141] The above specific embodiments do not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present disclosure shall be included within the protection scope of the present disclosure.

Claims

1. An operation and maintenance method based on a large language model, comprising: The following operations are performed through agent iteration: Determine the operation and maintenance results of the links corresponding to the completed links in the operation and maintenance tasks; By using a large language model, according to the link operation and maintenance results corresponding to the completed link, a target tool for continuing to perform the operation and maintenance task is determined from a tool set; The target tool is called to generate a link operation and maintenance result corresponding to the next link in the operation and maintenance task.

2. The method according to claim 1, wherein: The step of determining, through the large language model, a target tool for continuing to perform the operation and maintenance task from a tool set according to the operation and maintenance results of the links corresponding to the completed links, includes: The target tool is determined from the tool set through the large language model according to the model prompt data representing the operation and maintenance task and the link operation and maintenance results corresponding to the completed links.

3. The method according to claim 2, wherein: The step of determining the target tool from the tool set by using the large language model according to the model prompt data representing the operation and maintenance task and the link operation and maintenance results corresponding to the completed links includes: The target tool is determined from the tool set through the large language model according to the model prompt data, the link operation and maintenance results corresponding to the completed link, and the configuration data of the tools in the tool set.

4. The method according to claim 3, wherein: The configuration data includes tool description data and tool parameter data, and The step of determining the target tool from the tool set by using the large language model according to the model prompt data, the link operation and maintenance results corresponding to the completed link, and the configuration data of the tools in the tool set includes: Through the large language model, according to the model prompt data, the link operation and maintenance results corresponding to the completed link, and the tool description data and tool parameter data of the tools in the tool set, the target tool is determined from the tool set, and the target parameter value corresponding to the tool parameter data of the target tool is determined.

5. The method according to claim 4, wherein: The calling of the target tool to generate a link operation and maintenance result corresponding to the next link in the operation and maintenance task includes: By adopting the target tool of the target parameter value, the link operation and maintenance result corresponding to the next link in the operation and maintenance task is generated.

6. The method according to claim 5, wherein: The step of generating a link operation and maintenance result corresponding to a next link in the operation and maintenance task by using the target tool of the target parameter value includes: Determining a target data source according to the target tool and the target parameter value; By adopting the target tool of the target parameter value, the link operation and maintenance result corresponding to the next link in the operation and maintenance task is generated according to the data in the target data source.

7. The method according to claim 6, wherein: The target tool using the target parameter value generates, according to the data in the target data source, a link operation and maintenance result corresponding to the next link in the operation and maintenance task, including: During the process of processing the data in the target data source by the target tool adopting the target parameter value, the large language model is combined to perform auxiliary analysis and reasoning to generate the link operation and maintenance result corresponding to the next link in the operation and maintenance task.

8. The method according to claim 3, wherein: Also includes: The configuration data of the tools in the tool set are set by at least one of a model context protocol, a configuration file, and a database.

9. The method according to claim 1, wherein: The link operation and maintenance results corresponding to the completed links in the operation and maintenance task are determined, including: Determine the link operation and maintenance results corresponding to the preset number of completed links in the operation and maintenance task up to the present.

10. The method according to any one of claims 1 to 9, wherein: Also includes: During the execution of the operation and maintenance task, in response to the number of tool calls up to now exceeding a preset number threshold, a total operation and maintenance result is generated according to the link operation and maintenance results corresponding to the completed links.

11. The method according to any one of claims 1 to 9, wherein: Also includes: In response to the large language model, according to the link operation and maintenance results corresponding to the completed links, it is determined that the target tool does not exist in the tool set, and a total operation and maintenance result is generated according to the link operation and maintenance results corresponding to the completed links.

12. The method according to any one of claims 1 to 9, wherein: Also includes: In response to completing the operation and maintenance task, the overall operation and maintenance result of the operation and maintenance task is determined according to the link operation and maintenance result corresponding to the last link of the operation and maintenance task.

13. The method according to any one of claims 1 to 9, wherein: Also includes: Calling a first output tool to generate and display total result data in a first specified format according to the total operation and maintenance result of the operation and maintenance task; and / or The second output tool is called to generate and display the link result data in a second specified format according to the link operation and maintenance results.

14. An operation and maintenance device based on a large language model, comprising the following processing unit arranged in an intelligent agent to perform iterative operations: A result determination unit is configured to determine the link operation and maintenance results corresponding to the completed links in the operation and maintenance task; A tool determination unit is configured to determine, through a large language model, a target tool from a tool set for continuing to perform the operation and maintenance task according to the operation and maintenance results of the links corresponding to the completed links; The result generating unit is configured to call the target tool and determine the link operation and maintenance result corresponding to the next link in the operation and maintenance task.

15. An electronic device, characterized in that: include: at least one processor; as well as 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 the method according to any one of claims 1 to 13.

16. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 13.

17. A computer program product comprising: A computer program which, when executed by a processor, implements the method according to any one of claims 1 to 13.

Images