Intelligent agent service generation method and computer equipment

Abstract

The embodiment of the invention provides an agent service generation method and computer equipment, and is applied to an agent service development scene. The agent service generation method comprises the following steps: displaying a visual editing interface; obtaining pre-stored initialization data; according to an operation instruction for a management event in the agent service editing panel, an operation instruction for an assembly event in the agent service editing panel and the initialization data, the preview panel displays the assembly event of the agent; and generating a target agent service according to the assembly event of the agent displayed by the preview panel. According to the embodiment of the invention, unified management of the multi-manufacturer large language model, the MCP service and the cue word can be realized, and the code-free combined generation agent service can be realized, so that the defects of serious hard code coupling, configuration dynamics deficiency, insufficient hot update mechanism and over-high operation threshold of the construction scheme of the existing agent system can be solved.

Description

Technical Field

[0001] This application relates to the field of Internet technology, specifically to a method for generating intelligent agent services and a computer device. Background Technology

[0002] In related technologies, the construction of intelligent agent systems mainly relies on developers manually hard-coding the integration of various large language model services (such as OpenAI, Ollam, etc.) and context protocol services (MCP services).

[0003] The aforementioned intelligent agent system construction scheme suffers from severe hard-coded coupling, lack of dynamic configuration, insufficient hot-update mechanisms, and excessively high operational barriers. The interface protocols of different large language models vary significantly (e.g., API structure, authentication methods), requiring developers to repeatedly write adaptation code for each intelligent agent, resulting in high development costs, poor flexibility, and severe hard-coded coupling. The configuration process for prompts and MCP services must be implemented through code, making it impossible to dynamically adjust parameters or update strategies at runtime, leading to low maintenance efficiency and a lack of dynamic configuration. Adding or modifying intelligent agent configurations (e.g., model switching, service access) requires service restarts, interrupting business continuity and failing to achieve real-time hot-update capabilities, resulting in insufficient hot-update mechanisms. The lack of a visual configuration interface makes it difficult for ordinary users to participate in intelligent agent construction, relying on manual coding by professional developers, limiting technology dissemination and application innovation, and resulting in excessively high operational barriers. These shortcomings constrain the standardization and large-scale development of intelligent agent development; the industry urgently needs unified access standards and visual configuration schemes to improve efficiency and lower the barriers to entry. Summary of the Invention

[0004] This application provides a method and computer device for generating intelligent agent services, which can realize unified management of multi-vendor large language models, MCP services and prompt words, as well as code-free combination generation of intelligent agent services, solving the defects of current intelligent agent system construction schemes such as serious hard-coded coupling, lack of dynamic configuration, insufficient hot update mechanism and high operation threshold.

[0005] The method for generating an intelligent agent service according to an embodiment of this application includes: displaying a visual editing interface, the visual editing interface including an intelligent agent service editing panel and a preview panel; acquiring pre-stored initialization data, the initialization data originating from real samples of the intelligent agent service; displaying intelligent agent prompt word management events, large model service management events, and management model context protocol service management events on the preview panel according to operation instructions for management events in the intelligent agent service editing panel; displaying intelligent agent assembly events on the preview panel according to operation instructions for assembly events in the intelligent agent service editing panel, the intelligent agent prompt word management events, the large model service management events, the management model context protocol service management events, and the initialization data; and generating a target intelligent agent service based on the intelligent agent assembly events displayed on the preview panel.

[0006] In some implementations, the step of obtaining pre-stored initialization data, which originates from real samples of the agent service, includes: obtaining large language model service registration information, model context protocol service configuration information, prompt word template information, and agent configuration information from real samples of the agent service through a database; registering the large language model service registration information, model context protocol service configuration information, prompt word template information, and agent configuration information to the runtime environment and verifying them; and pre-constructing an agent instance pool using the verified large language model service registration information, model context protocol service configuration information, prompt word template information, and agent configuration information, wherein the data in the agent instance pool is determined as the initialization data.

[0007] In some implementations, registering the large language model service registration information, the model context protocol service configuration information, the prompt word template information, and the agent configuration information to the runtime environment and verifying them includes: verifying the service availability of the large language model service registration information; verifying the service status of the model context protocol service configuration information; and verifying the template integrity of the prompt word template information.

[0008] In some implementations, the step of displaying the agent's prompt word management event, large model service management event, and management model context protocol service management event in the preview panel according to the operation command for the management event in the agent service editing panel includes: in response to the operation command for the prompt word management event in the agent service editing panel, the preview panel displays a prompt word management event editing box; in response to the operation command for the prompt word management event editing box, the preview panel displays events for adding prompt words, deleting prompt words, or changing prompt word attributes, wherein the prompt word attributes include prompt word name, prompt word description information, and prompt word content.

[0009] In some implementations, the step of displaying agent prompt word management events, large model service management events, and management model context protocol service management events in the preview panel according to operation instructions for management events in the agent service editing panel includes: in response to operation instructions for large model service management events in the agent service editing panel, the preview panel displays a large model service management event editing box; in response to operation instructions for the large model service management event editing box, the preview panel displays events for adding a large model service, deleting a large model service, or changing large model service attributes, wherein the large model service attributes include large model service name, large model service address, key, type, model name, temperature, sampling example, and large model service description information.

[0010] In some implementations, the step of displaying agent prompt word management events, context protocol service management events, and management model context protocol service management events in the preview panel according to operation instructions for management events in the agent service editing panel includes: in response to operation instructions for context protocol service management events in the agent service editing panel, the preview panel displays a context protocol service management event editing box; in response to operation instructions for the context protocol service management event editing box, the preview panel displays events for adding a context protocol service, deleting a context protocol service, or changing context protocol service attributes, wherein the context protocol service attributes include the context protocol service name, Uniform Resource Locator address, transport protocol, and context protocol service description information.

[0011] In some implementations, the step of displaying the agent assembly event in the preview panel based on the operation instruction for the assembly event in the agent service editing panel, the agent's prompt word management event, the large model service management event, the management model context protocol service management event, and the initialization data includes: in response to the operation instruction for the assembly event in the agent service editing panel, the preview panel displays an agent assembly event editing box; in response to the operation instruction for the agent assembly event editing box, selecting a target prompt word from the agent's prompt word management event and the initialization data; in response to the operation instruction for the agent assembly event editing box, selecting a target large model service from the large model service management event and the initialization data; in response to the operation instruction for the agent assembly event editing box, selecting a target context protocol service from the context protocol service management event and the initialization data; and the preview panel displays the target prompt word, the target large model service, and the target context protocol service for assembling the agent.

[0012] In some embodiments, the method for generating the agent service further includes: storing the agent's prompt word management event, large model service management event, management model context protocol service management event, and assembly event in the preview panel into a database of shared documents; and displaying the invoked prompt word management event, large model service management event, management model context protocol service management event, and assembly event in the preview panel according to the call instruction for the shared document in the agent service editing panel.

[0013] In some embodiments, the method for generating the agent service further includes: displaying the invoked agent service in the preview panel according to a call instruction for a shared document in the agent service editing panel; and displaying the running process of the invoked agent service in the preview panel according to an input instruction for the invoked agent service in the preview panel.

[0014] The computer device in this application embodiment includes a processor and a memory. The memory stores a computer program, and the processor executes the intelligent agent service generation method as described in any of the above embodiments by calling the computer program stored in the memory.

[0015] This application embodiment displays a visual editing interface, including an agent service editing panel and a preview panel. It acquires pre-stored initialization data derived from real samples of the agent service. Based on operation instructions for management events in the agent service editing panel, the preview panel displays the agent's prompt word management events, large model service management events, and management model context protocol service management events. Based on operation instructions for assembly events in the agent service editing panel, the preview panel displays the agent's assembly events, including prompt word management events, large model service management events, management model context protocol service management events, and initialization data. Based on the assembly events displayed in the preview panel, the target agent service is generated. This enables unified management of multi-vendor large language models, MCP services, and prompt words, as well as no-code assembly-style generation of agent services. It addresses the shortcomings of current agent system construction schemes, such as severe hard-coded coupling, lack of dynamic configuration, insufficient hot update mechanisms, and excessively high operational barriers. Attached Figure Description

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

[0017] Figure 1A schematic diagram illustrating the method for generating intelligent agent services provided in the embodiments of this application.

[0018] Figure 2 This is a schematic diagram of the visual editing interface provided in the embodiments of this application.

[0019] Figure 3 This is a schematic diagram of the intelligent agent service editing panel in the visual editing interface provided in the embodiments of this application.

[0020] Figure 4 This is a schematic diagram of the prompt word management event editing box in the visual editing interface provided in the embodiments of this application.

[0021] Figure 5 This is a schematic diagram of the large model service management event editing box in the visual editing interface provided in the embodiments of this application.

[0022] Figure 6 This is a schematic diagram of the context protocol service management event editing box in the visual editing interface provided in the embodiments of this application.

[0023] Figure 7 This is a schematic diagram of the intelligent agent assembly event editing box in the visual editing interface provided in the embodiments of this application.

[0024] Figure 8 A schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] This application provides a method for generating intelligent agent services and a computer device. Exemplarily, the method for generating intelligent agent services in this application can be executed by the computer device in this application, wherein the computer device can be a terminal or a server, etc. The terminal can be a smartphone, tablet, laptop, desktop computer, smart TV, smart speaker, wearable smart device, personal computer (PC), smart vehicle terminal, etc. The terminal can also include a client, which can be a video client, shopping application client, reading application client, browser client, or instant messaging client, etc. The server can be 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, content delivery network (CDN), and big data and artificial intelligence platforms. However, it is not limited to these. The terminal and the server can be directly or indirectly connected through wired or wireless communication methods, which is not limited in this application.

[0027] The embodiments of this application can be applied to business scenarios based on intelligent agent services in the Internet, and can realize unified management of multi-vendor large language models, MCP services and prompt words, as well as code-free combination generation of intelligent agent services, solving the defects of current intelligent agent system construction schemes such as serious hard-coded coupling, lack of dynamic configuration, insufficient hot update mechanism and high operation threshold.

[0028] The solutions provided in this application within the field of Internet technology are specifically illustrated through the following embodiments. These embodiments are described in detail below. It should be noted that the order of description of the following embodiments is not intended to limit the priority of the embodiments.

[0029] The following describes the method for generating intelligent agent services provided by exemplary embodiments of this application, in conjunction with the application scenarios described above and with reference to the accompanying drawings. It should be noted that the above application scenarios are only shown to facilitate understanding of the principles of this application, and the embodiments of this application are not limited in any way in this respect.

[0030] Figure 1 This is a flowchart illustrating the method for generating intelligent agent services provided in embodiments of this application. Figure 1 As shown, the specific implementation process of this method includes: Step 010: Display the visual editing interface, which includes an agent service editing panel and a preview panel.

[0031] Specifically, the visual editing interface 100 can be referenced. Figures 2 to 7 .in, Figure 2 This is a schematic diagram showing the overall display of the visual editing interface 100 provided in the embodiments of this application. Figure 3 This is a schematic diagram of the agent service editing panel 110 in the visual editing interface 100 provided in the embodiments of this application. Figure 4 This is a schematic diagram of the prompt word management event editing box 131 in the visual editing interface 100 provided in the embodiments of this application. Figure 5 This is a schematic diagram of the large model service management event editing box 132 in the visual editing interface 100 provided in the embodiments of this application. Figure 6 This is a schematic diagram of the context protocol service management event editing box 133 in the visual editing interface 100 provided in the embodiments of this application. Figure 7 This is a schematic diagram of the intelligent agent assembly event editing box 134 in the visual editing interface 100 provided in the embodiments of this application.

[0032] Step 020: Obtain the pre-stored initialization data, which comes from real samples of the intelligent agent service.

[0033] Specifically, the pre-stored initialization data is obtained from samples of actual running agent projects. For example, at system startup, initialization data that has been verified by real business operations and is required for building the agent service can be loaded from an SQLite database. All the obtained initialization data originates from samples of actual running agent projects.

[0034] In some embodiments, step 020: Obtain pre-stored initialization data, wherein the initialization data originates from real samples of the intelligent agent service, including: Step 021: Obtain the large language model service registration information, model context protocol service configuration information, prompt word template information, and agent configuration information from the real samples of the agent service through the database.

[0035] Specifically, upon system startup, verified real-world business sample data is retrieved from databases such as SQLite, including four core configurations: large language model service registration information, model context protocol service configuration information, prompt word template information, and agent configuration information. The large language model service registration information includes connection parameters such as model vendor, API address, and key (e.g., OpenAI GPT-4 or locally deployed Llama 3). The model context protocol service configuration information (MCP service configuration) includes the path, input / output format, and dependency chain of tool invocation protocols (e.g., Web query, data analysis API). The prompt word template library includes historically accumulated high-quality prompt strategies (e.g., customer service script templates, data extraction instructions). The agent workflow configuration includes predefined agent assembly logic (e.g., multi-agent collaboration rules, memory module binding relationships).

[0036] Step 022: Register the large language model service registration information, model context protocol service configuration information, prompt word template information, and agent configuration information to the runtime environment and verify them.

[0037] Specifically, the loaded configuration information is injected into the runtime environment and its availability is verified. If the verification fails, an alarm is triggered and the faulty module is isolated to ensure that only compliant configuration information enters the next process.

[0038] In some embodiments, registering and verifying the large language model service registration information, model context protocol service configuration information, prompt word template information, and agent configuration information to the runtime environment includes: verifying the service availability of the large language model service registration information; verifying the service status of the model context protocol service configuration information; and verifying the template integrity of the prompt word template information.

[0039] Specifically, the loaded configuration information is injected into the runtime environment, and service availability verification of the large language model service registration information, service status verification of the model context protocol service configuration information, and template integrity verification of the prompt word template information are performed in stages. During service availability verification of the large language model service registration information, standardized test requests (such as simplified inference tasks) can be initiated based on the registration information (API address, key, etc.) to verify the interface response timeliness and content compliance. During service status verification of the model context protocol service configuration information, the model context protocol (MCP) configuration (service path, API format, dependency chain) is checked. When verifying the template integrity of the prompt word template information, the syntax tree parsing engine scans for template placeholder closure and variable reference validity, and performs static analysis on logical conflicts.

[0040] Step 023: The verified large language model service registration information, model context protocol service configuration information, prompt word template information, and agent configuration information are used to pre-build an agent instance pool. The data in the agent instance pool is determined as the initialization data.

[0041] Specifically, based on verified configuration data, reusable agent instances are created in advance, standardizing the traditional hard-coded agent construction into a "configuration as code" paradigm, reusing real business data to avoid adaptation costs, and improving convenience through pre-instantiation.

[0042] Step 030: Based on the operation instructions for managing events in the agent service editing panel 110, the preview panel 120 displays the agent's prompt word management events, large model service management events, and management model context protocol service management events.

[0043] Specifically, such as Figure 2 As shown, in the visual editing interface 100, the agent service editing panel 110 and the preview panel 120 are set independently. Figure 3 As shown, the intelligent agent service editing panel 110 can contain events for editing items 1 to 3. These events can be set as management events, used to manage the configuration information of the intelligent agent service.

[0044] In some embodiments, step 030: according to the operation instructions for managing events in the agent service editing panel 110, the preview panel 120 displays the agent's prompt word management event, large model service management event, and management model context protocol service management event, including: Step 031: In response to the operation command for the prompt management event in the agent service editing panel 110, the preview panel 120 displays the prompt management event editing box 131.

[0045] Step 032: In response to the operation command of the prompt word management event editing box 131, the preview panel 120 displays events for adding prompt words, deleting prompt words, or changing prompt word attributes. The prompt word attributes include the prompt word name, prompt word description information, and prompt word content.

[0046] Specifically, in response to operation commands for prompt word management events in the intelligent agent service editing panel 110, the preview panel 120 displays a prompt word management event editing box 131. Operation commands can include adding, changing, or deleting the corresponding prompt word management events. Prompt word management events can be saved to the backend for modular management.

[0047] like Figure 4As shown, in the context of adding a prompt word management event, the prompt word attribute is a blank field to be edited. You can add a prompt word management event by setting the prompt word name, prompt word description information, and prompt word content in the prompt word management event editing box 131, or you can change the prompt word name, prompt word description information, and prompt word content to change the prompt word management event.

[0048] In some embodiments, step 030: according to the operation instructions for managing events in the agent service editing panel 110, the preview panel 120 displays the agent's prompt word management event, large model service management event, and management model context protocol service management event, including: Step 033: In response to the operation command for the large model service management event in the agent service editing panel 110, the preview panel 120 displays the large model service management event editing box 132; Step 034: In response to the operation command of the large model service management event edit box 132, the preview panel 120 displays events for adding a large model service, deleting a large model service, or changing the attributes of a large model service. The attributes of a large model service include the large model service name, large model service address, key, type, model name, temperature, sampling example, and large model service description information.

[0049] Specifically, in response to operation commands for large model service management events in the agent service editing panel 110, the preview panel 120 displays a large model service management event editing box 132. Operation commands can include adding, changing, or deleting the corresponding large model service management event. Large model service management events can be saved to the backend for modular management.

[0050] like Figure 5 As shown, in the "Add Large Model Service Management Event" section, the Large Model Service attributes are blank fields to be edited. You can add a Large Model Service Management Event by setting the Large Model Service Name, Large Model Service Address, Key, Type, Model Name, Temperature, Sampling Example (Top-p), and Large Model Service Description in the Large Model Service Management Event edit box 132. You can also modify the Large Model Service Management Event by changing these attributes.

[0051] In some embodiments, step 030: according to the operation instructions for managing events in the agent service editing panel 110, the preview panel 120 displays the agent's prompt word management event, large model service management event, and management model context protocol service management event, including: Step 035: In response to the operation command for the context protocol service management event in the agent service editing panel 110, the preview panel 120 displays the context protocol service management event editing box 133; Step 036: In response to the operation command for the context protocol service management event edit box 133, the preview panel 120 displays events for adding a context protocol service, deleting a context protocol service, or changing the context protocol service attributes. The context protocol service attributes include the context protocol service name, the Uniform Resource Locator address, the transport protocol, and the context protocol service description information.

[0052] Specifically, in response to operation commands for context protocol service management events in the agent service editing panel 110, the preview panel 120 displays a context protocol service management event editing box 133. Operation commands can include adding, changing, or deleting the corresponding context protocol service management event. Context protocol service management events can be saved to the backend for modular management.

[0053] like Figure 6 As shown, in the context protocol service management event, the context protocol service attribute is a blank item to be edited. You can add the context protocol service management event by setting the context protocol service name, Uniform Resource Locator (URL) address, transport protocol and context protocol service description information in the context protocol service management event editing box 133. You can also change the context protocol service name, Uniform Resource Locator (URL) address, transport protocol and context protocol service description information to change the context protocol service management event.

[0054] Step 040: Based on the operation instructions for the assembly event in the agent service editing panel 110, the agent's prompt word management event, the large model service management event, the management model context protocol service management event, and the initialization data, the preview panel 120 displays the agent's assembly event.

[0055] Specifically, such as Figure 3 As shown, the intelligent agent service editing panel 110 can contain an event for editing item 4. Editing item 4 can be set as an assembly event, and the intelligent agent service can be assembled by operating the assembly event.

[0056] In some embodiments, step 040: Based on the operation instructions for the assembly event in the agent service editing panel 110, the agent's prompt word management event, the large model service management event, the management model context protocol service management event, and the initialization data, the preview panel 120 displays the agent's assembly event, including: Step 041: In response to the operation command for the assembly event in the agent service editing panel 110, the preview panel 120 displays the agent assembly event editing box 134.

[0057] Specifically, such as Figure 7 As shown, in the agent assembly event editing box 134, the name, description information, prompt words, model, and context protocol service (MCP) are editable items. The agent assembly event can be set by setting the name, description information, prompt words, model, and context protocol service (MCP), and the agent assembly event can also be changed by changing the name, description information, prompt words, model, and context protocol service (MCP).

[0058] Step 042: In response to the operation command for the agent assembly event edit box 134, select the target prompt word from the agent's prompt word management event and initialization data.

[0059] Specifically, such as Figure 7 As shown, in the agent assembly event edit box 134, the prompt word is a selectable item. The initialization data may include prompt word templates loaded from the database. The selectable prompt words include prompt word templates loaded from the database and prompt word templates saved in previously edited prompt word management events. A target prompt word can be set in the selectable prompt words, and the set target prompt word is displayed in the agent assembly event edit box 134.

[0060] Step 043: In response to the operation command for the agent assembly event edit box 134, select the target large model service from the large model service management event and initialization data.

[0061] Specifically, such as Figure 7 As shown, in the agent assembly event edit box 134, the model is a selectable item. Initialization data may include large model service templates loaded from the database. The selectable items for the model include large model service templates loaded from the database and large model service templates saved in previously edited model management events. The target large model service can be set in the selectable items for the model, and the set target large model service is displayed in the agent assembly event edit box 134.

[0062] Step 044: In response to the operation command for the agent assembly event edit box 134, select the target context protocol service from the context protocol service management events and initialization data.

[0063] Specifically, such as Figure 7As shown, in the agent assembly event edit box 134, MCP is a selectable item. Initialization data may include context protocol service templates loaded from the database. The selectable items for MCP include context protocol service templates loaded from the database and context protocol service templates saved in previously edited model management events. In the selectable items for the model, the target context protocol service can be set, and the set target context protocol service is displayed in the agent assembly event edit box 134.

[0064] Step 045: Preview panel 120 displays the target cue words, target large model service, and target context protocol service used to assemble the agent.

[0065] Step 050: Generate the target agent service based on the agent assembly event displayed in the preview panel 120.

[0066] Specifically, after a user submits a new configuration, the backend injects the updated components into the runtime environment through memory reloading technology to generate the target intelligent agent service. This enables unified management of multi-vendor large language models, MCP services and prompt words, as well as code-free assembly generation of intelligent agent services. It solves the defects of current intelligent agent system construction schemes, such as severe hard-coded coupling, lack of dynamic configuration, insufficient hot update mechanism and high operation threshold.

[0067] In some implementations, the method for generating agent services further includes: Store the agent's prompt word management events, large model service management events, management model context protocol service management events, and assembly events in the preview panel 120 into the shared document database; Based on the call command for the shared document in the intelligent agent service editing panel 110, the preview panel 120 displays the called prompt word management event, large model service management event, management model context protocol service management event, and assembly event.

[0068] Specifically, such as Figure 3 As shown, the intelligent agent service editing panel 110 can contain an event for editing item 5. Editing item 5 can be set as a shared document call event. The relevant data displayed in the preview panel 120 can be stored in the shared document database. The shared document database can be called through editing item 5.

[0069] Specifically, after configuring the agent through the visual editing interface 100, the system can automatically, or the user can trigger a storage command, store prompt word management events, large model service management events, management model context protocol service management events, and assembly events into the shared document's database. Upon triggering a call command for the shared document in the agent service editing panel 110, the system loads the configuration from the database in real time and dynamically renders the status of each event in the preview panel 120. Thus, after successful agent creation, users can directly call the agent through an interface or interact with it through a front-end interface, supporting context storage and suitable for various intelligent application scenarios.

[0070] In some implementations, the method for generating agent services further includes: Based on the call command for the shared document in the agent service editing panel 110, the preview panel 120 displays the invoked agent service.

[0071] Specifically, when the instruction to call the shared document is triggered, the system responds according to the following logic and presents it visually in the preview panel 120. Through configuration loading, it extracts four types of pre-stored event information from the database of the shared document: prompt word management event, large model service management event, management model context protocol service management event, and assembly event. Through context association, it automatically restores the intelligent agent's running environment and dynamically renders the called component to the preview panel 120.

[0072] Based on the input command for the agent service invoked in the preview panel 120, the preview panel 120 displays the running process of the invoked agent service.

[0073] Specifically, after a user inputs a command through the preview panel 120 (such as the test statement "check Beijing weather"), the system deconstructs the intelligent agent's operation process in real time: it automatically generates the corresponding result in response to the input command, and subsequent requests can carry the generated result to maintain the context (such as continuously asking "trip planning for Beijing"), which will respond to Beijing weather to plan a trip to Beijing.

[0074] In some embodiments, this application also provides a computer device 200, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0075] Figure 8 A schematic structural diagram of the computer device provided in the embodiments of this application, such as Figure 8As shown, the computer device 200 may include: a communication interface 201, a memory 202, a processor 203, and a communication bus 204. The communication interface 201, memory 202, and processor 203 communicate with each other via the communication bus 204. The communication interface 201 is used for data communication between the computer device 200 and external devices. The memory 202 can be used to store software programs and modules, and the processor 203 runs the software programs and modules stored in the memory 202, such as the software programs for the corresponding operations in the aforementioned method embodiments.

[0076] In some embodiments, the processor 203 may invoke software programs and modules stored in the memory 202 to execute the above-described method for generating intelligent agent services.

[0077] In some embodiments, the computer device 200 may be integrated into a terminal or server that has storage and a processor and thus computing power, or the computer device 200 may be the terminal or server.

[0078] This application also provides a computer-readable storage medium for storing a computer program. This computer-readable storage medium can be applied to a computer device, and the computer program causes the computer device to execute the corresponding processes in the methods described above in the embodiments of this application; for brevity, further details are omitted here.

[0079] This application also provides a computer program product including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the corresponding processes in the methods described above in the embodiments of this application. For brevity, these details will not be elaborated further here.

[0080] This application also provides a computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the corresponding processes in the methods described above in the embodiments of this application. For brevity, these details will not be elaborated further here.

[0081] It should be understood that the processor in the embodiments of this application may be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above method embodiments can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor described above can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0082] It is understood that the memory in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DRRAM). It should be noted that the memory used in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0083] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0084] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0085] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units 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.

[0086] In this application embodiment, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0087] The units described 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.

[0088] In addition, the functional units in the 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.

[0089] If a function 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 a 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 several instructions to cause a computer device (which may be a personal computer or a server) 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 program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0090] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for generating intelligent agent services, characterized in that, The method for generating the intelligent agent service includes: Display a visual editing interface, which includes an agent service editing panel and a preview panel; Obtain pre-stored initialization data, which is derived from real samples of the intelligent agent service; Based on the operation instructions for managing events in the agent service editing panel, the preview panel displays the agent's prompt word management events, large model service management events, and management model context protocol service management events; Based on the operation instructions for the assembly event in the agent service editing panel, the agent's prompt word management event, the large model service management event, the management model context protocol service management event, and the initialization data, the preview panel displays the agent's assembly event; Based on the assembly events of the agent displayed in the preview panel, generate the target agent service.

2. The method for generating intelligent agent services as described in claim 1, characterized in that, The process of acquiring pre-stored initialization data, wherein the initialization data originates from real samples of the intelligent agent service, includes: The database is used to obtain the large language model service registration information, model context protocol service configuration information, prompt word template information, and intelligent agent configuration information from real samples of intelligent agent services. Register the large language model service registration information, the model context protocol service configuration information, the prompt word template information, and the agent configuration information to the runtime environment and verify them; The verified large language model service registration information, model context protocol service configuration information, prompt word template information, and agent configuration information are used to pre-build an agent instance pool, and the data in the agent instance pool is determined as the initialization data.

3. The method for generating intelligent agent services as described in claim 2, characterized in that, The step of registering and verifying the large language model service registration information, the model context protocol service configuration information, the prompt word template information, and the agent configuration information to the runtime environment includes: Verify the service availability of the large language model service registration information; Verify the service status of the model context protocol service configuration information; Verify the integrity of the prompt word template information.

4. The method for generating intelligent agent services as described in claim 1, characterized in that, Based on the operation instructions for managing events in the agent service editing panel, the preview panel displays the agent's prompt word management events, large model service management events, and management model context protocol service management events, including: In response to an operation command for a prompt management event in the agent service editing panel, the preview panel displays a prompt management event editing box; In response to the operation command of the prompt word management event editing box, the preview panel displays events for adding prompt words, deleting prompt words, or changing prompt word attributes. The prompt word attributes include prompt word name, prompt word description information, and prompt word content.

5. The method for generating intelligent agent services as described in claim 1, characterized in that, Based on the operation instructions for managing events in the agent service editing panel, the preview panel displays the agent's prompt word management events, large model service management events, and management model context protocol service management events, including: In response to an operation command for the large model service management event in the agent service editing panel, the preview panel displays the large model service management event editing box; In response to the operation command for the large model service management event editing box, the preview panel displays events for adding a large model service, deleting a large model service, or changing the attributes of a large model service. The attributes of the large model service include the large model service name, large model service address, key, type, model name, temperature, sampling example, and large model service description information.

6. The method for generating intelligent agent services as described in claim 1, characterized in that, Based on the operation instructions for managing events in the agent service editing panel, the preview panel displays the agent's prompt word management events, context protocol service management events, and management model context protocol service management events, including: In response to an operation command for a context protocol service management event in the agent service editing panel, the preview panel displays a context protocol service management event editing box; In response to the operation command for the context protocol service management event editing box, the preview panel displays events for adding a context protocol service, deleting a context protocol service, or changing the context protocol service attributes. The context protocol service attributes include the context protocol service name, the Uniform Resource Locator address, the transport protocol, and the context protocol service description information.

7. The method for generating intelligent agent services as described in claim 1, characterized in that, The assembly event of the agent is displayed in the preview panel based on the operation instructions for the assembly event in the agent service editing panel, the agent's prompt word management event, the large model service management event, the management model context protocol service management event, and the initialization data, including: In response to an operation command for an assembly event in the agent service editing panel, the preview panel displays an agent assembly event editing box; In response to an operation command for the assembly event edit box of the agent, a target prompt word is selected from the prompt word management event of the agent and the initialization data; In response to the operation command for the intelligent agent assembly event edit box, a target large model service is selected from the large model service management event and the initialization data; In response to the operation command for the intelligent agent assembly event edit box, a target context protocol service is selected from the context protocol service management event and the initialization data; The preview panel displays the target cue words, the target large model service, and the target context protocol service used to assemble the agent.

8. The method for generating intelligent agent services as described in claim 1, characterized in that, The method for generating the intelligent agent service further includes: Store the agent's prompt word management events, large model service management events, management model context protocol service management events, and assembly events in the preview panel into the shared document database; Based on the call command for the shared document in the intelligent agent service editing panel, the preview panel displays the called prompt word management event, large model service management event, management model context protocol service management event, and assembly event.

9. The method for generating intelligent agent services as described in claim 1, characterized in that, The method for generating the intelligent agent service further includes: Based on the invocation command for the shared document in the agent service editing panel, the preview panel displays the invoked agent service; Based on the input command for the agent service invoked in the preview panel, the preview panel displays the running process of the invoked agent service.

10. A computer device, comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to execute the method for generating intelligent agent services as described in any one of claims 1 to 9 through the computer program.

Images