Methods, apparatus, media, and computer program products for creating smart components

Abstract

A method, device, medium and computer program product for creating a smart component are provided. The method comprises receiving a first operation of displaying one or more historical information; receiving a second operation of selecting at least one first historical information, wherein the at least one first historical information belongs to the one or more historical information and is related to the creation of a first smart component; and receiving a third operation of creating a second smart component based on at least the at least one first historical information. The method can reuse historical information as a reference for generating a new smart component, without requiring the user to manually reorganize and describe the complete task requirements and execution process, which reduces the cost of manual description, simplifies the component creation operation process, ensures that the newly generated component is adapted to the actual business scenario, and improves the fit and usability of the component generation result.

Description

Technical Field

[0001] One or more of the above relate to methods for creating intelligent components, devices for creating intelligent components, and corresponding electronic devices, computer-readable storage media, and computer program products. Background Technology

[0002] With the rapid iteration and widespread application of Large Language Model (LLM) technology, intelligent agent systems with natural language understanding and autonomous content generation capabilities have become widely popular. LLM can accurately parse user-input natural language commands and generate text content, structured data, and task execution plans that meet specific needs based on training knowledge and logical reasoning abilities, providing core technical support for scenarios such as intelligent interaction and automated task processing. In the LLM-based agent system architecture, the agent's core execution capabilities are mainly realized through two types of core components: Skill functional components and SubAgent sub-intelligent system components. These two types of components are the core carriers for the agent to complete various complex tasks and respond to diverse user needs. User interaction with the intelligent agent is conducted in sessions, and all session data is completely recorded by the system platform. The retained session content comprehensively covers the user's original task requirements, the agent's entire execution process, and the detailed process of various tool calls, fully reconstructing the interaction and execution logic of each task. Currently, the mainstream conversational component creation solutions in the industry mainly rely on real-time natural language dialogue between users and agents to drive large language models to automatically generate the required Skill functions or SubAgent sub-intelligent system components. This eliminates the need for manual pre-writing of complete configuration rules, offering a certain degree of convenience and becoming the mainstream method for lightweight creation of Agent components.

[0003] However, existing agent component creation schemes based on natural language dialogue still have many substantial technical defects in practical applications, which seriously restrict the creation efficiency, accuracy and iterability of Skill and SubAgent components, and cannot meet the needs of customized and refined creation of intelligent agent components in complex scenarios. Summary of the Invention

[0004] This summary section is provided to provide a brief overview of the concepts, which will be described in detail in the detailed description section below. This summary section is not intended to identify key or essential features of the claimed invention, nor is it intended to limit the scope of the claimed invention.

[0005] At least one scenario of this document provides a method for creating a smart component, comprising: receiving a first operation to display one or more historical information; receiving a second operation to select at least one first historical information, wherein the at least one first historical information belongs to one or more historical information and is related to the creation of a first smart component; and receiving a third operation to create a second smart component based at least on at least one first historical information.

[0006] At least one other embodiment of this document provides an apparatus for creating a smart component, comprising: a first operation unit configured to receive a first operation to display one or more historical information; a second operation unit configured to receive a second operation to select at least one first historical information, wherein the at least one first historical information belongs to one or more historical information and is related to the creation of a first smart component; and a third operation unit configured to receive a third operation to create a second smart component based at least on at least one first historical information.

[0007] At least one further embodiment of this document provides an electronic device comprising: at least one processor, and a memory storing one or more computer program instructions; wherein the one or more computer program instructions are executed by the at least one processor to perform the method provided in at least one embodiment of this document.

[0008] At least one further embodiment of this document provides a non-transitory computer-readable storage medium for storing computer-readable instructions, wherein the method provided in at least one embodiment of this document is implemented when the computer-readable instructions are executed by a processor.

[0009] At least one embodiment of this document provides a computer program product comprising one or more computer instructions, wherein the one or more computer instructions are executed by a processor to perform the method provided in at least one embodiment of this document.

[0010] This paper provides methods for creating intelligent components, along with corresponding devices, media, and products, for one or more scenarios. The methods presented here can reuse historical information (e.g., valid historical session data verified in real-world scenarios) as a reference for generating new Skills and sub-intelligent system components. This eliminates the need for users to manually reconstruct and verbally describe complete task requirements and execution processes, reducing manual description costs and simplifying the component creation process. Simultaneously, it avoids the discrepancy between manually recounting requirements and actual historical execution scenarios, ensuring that newly generated components are compatible with actual business scenarios and improving the fit and usability of the generated components. Attached Figure Description

[0011] The above and other features, advantages, and aspects of the various scenarios herein will become more apparent when taken in conjunction with the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0012] Figure 1 The illustrations depict application scenarios of methods and devices for creating intelligent components provided in at least one of the cases described herein.

[0013] Figure 2 The flowchart illustrating a method for creating intelligent components provided in at least one scenario of this paper is shown schematically;

[0014] Figures 3 to 5 The illustrations provide a schematic diagram related to the session context extraction phase for at least one scenario presented in this paper;

[0015] Figures 6 to 7 The illustrations provide a diagram relating to the requirements analysis and generation phases for at least one scenario presented in this paper.

[0016] Figures 8 to 9 The illustrations provide a diagram relating to the dialogue revision phase for at least one scenario presented in this paper.

[0017] Figures 10 to 11 The illustrations provide a schematic diagram related to the confirmation and preservation phase for at least one scenario described herein;

[0018] Figure 12 The flowchart illustrating a portion of the process for creating a smart component is shown in at least one scenario described herein.

[0019] Figure 13 The schematic diagram illustrates a structural block diagram of a device 1300 for creating intelligent components, provided in at least one scenario of this document;

[0020] Figure 14 A schematic block diagram of an electronic device 1400 in at least one of the present invention is shown. Detailed Implementation

[0021] The present invention will now be described in more detail with reference to the accompanying drawings. While some aspects of the present invention are illustrated in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the aspects set forth herein; rather, these aspects are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the accompanying drawings and aspects are for illustrative purposes only and are not intended to limit the scope of this invention.

[0022] It should be understood that the steps described in the method embodiments herein may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this document is not limited in this respect.

[0023] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "according to" means "at least partially according to". The term "one situation" means "at least one situation"; the term "another situation" means "at least one additional situation"; the term "some situations" means "at least some situations". Definitions of other terms will be given in the following description.

[0024] It should be noted that the concepts of "first" and "second" mentioned in this article are only used to distinguish different models, devices, modules or units, and are not used to limit the order of the functions performed by these models, devices, modules or units or their interdependencies.

[0025] It should be noted that the terms "one" and "multiple" used in this document are illustrative rather than restrictive, and those skilled in the art should understand that, unless explicitly stated in the context, they should be interpreted as "one or more".

[0026] The names of messages or information exchanged between multiple devices, modules or units in the embodiments herein are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0027] It is understood that the data involved in this article (including but not limited to the data itself, the acquisition, use, storage or deletion of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0028] It is understood that before using the technical solutions disclosed in each scenario in this article, relevant users should be informed of the type, scope of use, and usage scenarios of the information involved in this article and their authorization should be obtained through appropriate means in accordance with relevant laws and regulations. Relevant users may include any type of rights holder, such as individuals, enterprises, or groups.

[0029] For example, in the process of creating smart components, which will be introduced later, historical information and operational information may be involved. When it is necessary to inform relevant users in accordance with relevant laws and regulations, a prompt message will be sent to the relevant users to clearly inform them, allowing them to choose whether to allow the use of such historical information and operational information based on the prompt message. This prompt can be in the form of a pop-up window, which can present the prompt message in text format. In addition, the pop-up window can also include a selection control for users to choose "agree" or "disagree".

[0030] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation method described in this article. Other methods that comply with relevant laws and regulations may also be applied to the implementation method described in this article.

[0031] Current mainstream conversational Skill and SubAgent component creation solutions rely entirely on natural language interactive dialogue between users and intelligent agents to generate and iteratively optimize components. The overall creation process depends on real-time dialogue interaction and the dynamic generation capabilities of large language models. The specific execution steps are as follows:

[0032] First, users can initiate a component creation request by describing the functional requirements, application scenarios, and core execution logic of the Skill functional component or SubAgent sub-intelligent system to be created in natural language through the system's preset component creation entry point.

[0033] Second, after receiving the user's creation instruction, the intelligent agent calls the Large Language Model (LLM) to perform semantic parsing and intent recognition on the user's natural language requirements, determines the type of component to be created, and breaks down and sorts out the various configurations required by the component. If the current user's requirements information is missing, ambiguous, or unclear, the agent will proactively ask the user follow-up questions to supplement and improve the key information.

[0034] Third, the large language model automatically generates the corresponding component configuration content based on the user's initial needs and the complete information supplemented in the dialogue, and outputs it to the dialogue interaction link in plain text form for the user to preview and verify the generation effect in real time.

[0035] Fourth, users provide feedback on the previewed component text configuration content using natural language, expressing their suggestions for modification, optimization needs, or functional adjustments. The large language model regenerates the entire content based on the complete content generated in the previous round, combined with the user's modification instructions. The generated content is iteratively corrected through multiple rounds of dialogue until the user confirms that the component content meets their usage requirements.

[0036] Fifth, after the user confirms that the configuration content of the iterated component is correct, the system will write the final generated configuration content into the component definition file and complete the registration and storage of the component in the intelligent agent platform, so that it can be called and executed normally. However, if subsequent modifications such as parameter adjustment and function optimization are required for the registered Skill or SubAgent component, the user can only manually edit it by entering the platform's dedicated management page, and cannot rely on the dialogue interaction mode for synchronous optimization.

[0037] It is evident that existing agent component creation schemes based on natural language dialogue still suffer from numerous substantial technical shortcomings in practical applications. These shortcomings severely restrict the creation efficiency, accuracy, and iterability of Skill and SubAgent components, and fail to meet the customized and refined creation needs of intelligent agent components in complex scenarios. The main problems with existing solutions include the following aspects:

[0038] (1) The cost of obtaining the existing component generation basis is high and the system cannot reuse historical valid session data: the Skill or SubAgent creation task process that users have fully executed in past sessions and verified as feasible in actual scenarios cannot be directly used as the reference basis for generating new components; when creating a brand new component, users can only rely on manual memory to reorganize and verbally describe the task requirements and execution process. Not only is manual description costly and cumbersome, but the manual recitation of requirements often deviates from the actual historical execution scenario, which can easily lead to the newly created component not matching the actual application scenario.

[0039] (2) The existing component generation results are unstructured and do not support fine-grained editing and adjustment: The final output of Skill and SubAgent components in the existing solution are all in plain text format, without standardized configuration fields and structured data architecture, making it impossible to accurately locate and independently modify a single configuration item or functional field; when it is necessary to adjust the component function locally, all content can only be regenerated as a whole through the large language model, the modification method is rough, the adjustment precision is low, the component iteration process is uncontrollable, and new content deviations are very easy to occur.

[0040] (3) The operation mode is fragmented and cannot be flexibly adapted: During the dynamic process of component creation, the generated content can only be adjusted through natural language dialogue. If manual precise editing and optimization are required, it must be done after the component is fully created. The data status and editing progress of the two optimization modes, dialogue modification and manual editing, cannot be communicated. The two operation modes are independent of each other and cannot be linked, which reduces the flexibility and efficiency of component creation and optimization.

[0041] (4) The component creation process does not support dynamic supplementation of references: After the user initiates the component creation task, the generation logic only relies on the interaction content of the current dialogue. It is not possible to retrieve or reference other related historical conversation data at any time to correct, supplement and expand the generated content. If the generated content is found to have defects or missing functions, the current task can only be terminated and a completely new component creation process can be initiated. The overall task redundancy is high and the iteration efficiency is low.

[0042] To at least partially address the aforementioned technical problem, at least one aspect of this document provides a method for creating a smart component, comprising: receiving a first operation to display one or more historical information; receiving a second operation to select at least one first historical information, wherein the at least one first historical information belongs to one or more historical information and is related to the creation of a first smart component; and receiving a third operation to create a second smart component based at least on at least one first historical information.

[0043] In conjunction with the methods described above for at least one of the embodiments herein, one or more embodiments herein also provide a corresponding device, computer-readable storage medium, and computer program product.

[0044] The following detailed description, with reference to the accompanying drawings, illustrates one or more scenarios and some examples thereof.

[0045] Figure 1 The illustrations depict application scenarios of methods and devices for creating intelligent components, as provided in at least one of the cases presented herein.

[0046] like Figure 1 As shown, in this application scenario 100, a user 101, a terminal device 102, a communication network 103, and a server 104 may be included.

[0047] For example, user 101 may include a user or other relevant party using an application for creating smart components. The application for creating smart components may include, but is not limited to, web applications (such as browser-based smart component creation applications), mobile terminal applications (Android applications, iOS applications), etc., which provide the client described herein; user 101 may initiate a request to create a smart component on the client. This client may run on terminal device 102. Terminal device 102 may load and execute an executable program or instructions that run the client, thereby enabling user 101 to initiate a request to create a smart component using the client; this is not limited in the context of this document.

[0048] For example, terminal device 102 may include any electronic device capable of running the aforementioned client. This arbitrary electronic device can provide users with an interactive page for operation and use. Such arbitrary electronic devices include, but are not limited to, mobile terminals, mobile phones, tablets, portable computers, desktop computers, smart wearable devices, smart home appliances, or smart vehicle terminals.

[0049] For example, communication network 103 may include any suitable wired or wireless communication network. Wired communication may include, but is not limited to, twisted-pair communication, coaxial cable communication, and fiber optic communication. Wireless communication may include, but is not limited to, cellular mobile communication, Wi-Fi communication, satellite communication, wireless local area network, and wireless wide area network.

[0050] For example, server 104 runs a server program corresponding to the client and can communicate with terminal device 102 via wired or wireless communication networks. Server 104 can be, for example, a local area network server, a wide area network server, a cloud server, etc., and specifically can be an enterprise-level application server, a distributed server cluster, or a cloud server. Furthermore, in some cases, server 104 and terminal device 102 can be integrated into one unit; in this case, the aforementioned communication network 103 may not be necessary. For example, in response to user 101's interactive operation on the client's interactive page, the client on terminal device 102 can obtain the corresponding operation related to creating smart components (such as receiving user natural language input, displaying historical information, displaying configuration items, and receiving user operations (such as the first operation and the second operation, which will be introduced later)). Then, the client on terminal device 102 can send the information corresponding to the relevant operation to the server on server 104, so that the server on server 104 can perform subsequent processing such as calling model parsing and searching historical information based on the obtained information. Finally, the server on server 104 can present the content corresponding to the processing to the user (such as displaying historical information, displaying smart component information, and displaying configuration items, which will be introduced later) so that the user can perform further operations.

[0051] For example, the methods for creating intelligent components provided in at least one of the scenarios described herein can be implemented in any suitable manner, such as in software.

[0052] The terminal device 102 may include, for example, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a Digital Signal Processor (DSP), a Neural Processing Unit (NPU), or other processing units with data processing and / or instruction execution capabilities, as well as corresponding storage units (e.g., shared cache, memory, etc.). For example, the CPU may use the x86 instruction set, ARM instruction set, RISC-V instruction set, etc.

[0053] The terminal device 102 can be equipped with an operating system, application programming interface (API), etc., and can run the client by running executable programs or instructions.

[0054] For example, terminal device 102 may also include an output component, such as a display device, which may be a liquid crystal display (LCD), an organic light-emitting diode (OLED) display, a quantum dot light-emitting diode (QLED) display, or a light-emitting diode display (e.g., a micro-LED display), etc., without limitation herein. For example, the display component may display an interactive page, allowing a user to interact with information on a client on terminal device 102 through the interactive interface.

[0055] Figure 2 The flowchart illustrating a method for creating intelligent components provided in at least one scenario of this paper is shown schematically.

[0056] like Figure 2 As shown, method 200 in this scenario may include steps S210 to S230. For example, the executing entity of method 200 may be a smart agent, which can be used to create smart components, invoke already created smart components, and perform other operations. This smart agent may be deployed on a server (as described above). Figure 1 In the aforementioned server 104, the requirements parsing, LLM invocation, component generation, multi-round iterative processing, and component registration, which will be described later, are all completed by the server-side Agent. This intelligent component may include Skill functional components, SubAgent sub-intelligent system components, etc.

[0057] Reference Figure 2In step S210, the first operation can be received, and one or more historical information can be displayed.

[0058] As an example, the first operation may include a first user (i.e., the user who wants to create the smart component) selecting or clicking a button or control on the client. Upon receiving this operation, one or more pieces of historical information may be displayed. As another example, the first operation may include a first user selecting or clicking a button or control on the client via voice or other instructions, such as "Please click the button to display historical information"; upon receiving this operation, one or more pieces of historical information may be displayed.

[0059] As an example, displaying one or more historical information entries can include displaying them as a list. For each historical entry in the list, the displayed content can include: the name of the historical entry, a partial description, etc. For example, the name of the historical entry could include: Implementing a user authentication module, Refactoring the API interface, ai-bug-fixer (an official skill), etc.; the description could include: for the user authentication module implementation, "Help me design a user authentication module | It can be broken down as follows: Login: Account Password / SMS / Third-party, etc."; for the API interface refactoring, "Help me refactor the API interface, unify error codes | Suggest a unified response structure, etc."; for ai-bug-fixer, "It can automatically diagnose vulnerabilities in the code, generate fix patches, and provide exception handling and regression testing suggestions."

[0060] In addition to displaying one or more historical information entries in a list format, one or more other visual layout methods can be used, such as timelines, flow cards, tree directories, grouped collapsible panels, and tabbed pagination. This article does not impose any restrictions on this approach.

[0061] As an example, historical information may include historical conversations, historical documents, historical files, official skills, etc. This information may include the second user's historical needs (also known as historical intents); this information may be related to the creation of historical smart components, such as whether the second user previously intended to use or has already used this information to create historical smart components. The second user may be the same as the aforementioned first user; alternatively, the second user may be different from the aforementioned first user.

[0062] Continue to refer to Figure 2 In step S220, a second operation can be received to select at least one first historical information, wherein the at least one first historical information belongs to one or more historical information and is related to the creation of the first intelligent component.

[0063] As an example, the second operation may include the first user selecting or clicking historical information on the client or a button or control next to that historical information. As another example, the second operation may include the first user selecting or clicking historical information on the client or a button or control next to that historical information via voice or other instructions, such as the voice prompt saying "Please help me select historical information 'Implement user authentication module'".

[0064] Upon receiving the second operation, at least one first historical information can be selected; for example, the selection of the first historical information can be indicated by making the first historical information display a different color than the unselected historical information, such as making the selected first historical information gray or black; or, for another example, the selection of the first historical information can be indicated by changing the state of the control next to the first historical information, such as changing the state of the control from "referenced" to "referenced".

[0065] As an example, the at least one first historical information may include one first historical information. As another example, the at least one first historical information may include multiple first historical information.

[0066] As an example, this first historical information may be related to the creation of the first intelligent component, such as the first intelligent component that was intended to be used or has already been created in the past.

[0067] Continue to refer to Figure 2 In step S230, a third operation can be received, which creates a second intelligent component based on at least one first historical information.

[0068] As an example, the third operation may include the first user selecting or clicking a button or control on the client. Upon receiving this operation, a second smart component can be created based on at least one piece of first historical information. As another example, the third operation may include the first user selecting or clicking a button or control on the client via voice or other instructions, such as "Please click the button to create the second smart component." Upon receiving this operation, a second smart component can be created based on at least one piece of first historical information.

[0069] As an example, the second smart component can be different from the first smart component.

[0070] As can be seen, the methods provided in this paper, based on one or more scenarios, can reuse historical information (e.g., historically valid conversation data verified in actual scenarios) as a reference for generating new Skills and sub-intelligent system components. This eliminates the need for users to manually reconstruct and verbally describe complete task requirements and execution processes, reducing the cost of manual description and simplifying the component creation process. At the same time, it avoids the problem of discrepancies between manual recitation of requirements and real historical execution scenarios, ensuring that newly generated components are adapted to actual business scenarios and improving the fit and usability of component generation results.

[0071] According to at least one of the embodiments described herein, method 200 may further include: receiving an operation related to at least one first historical information and displaying first smart component information.

[0072] As an example, after selecting at least one piece of first historical information, this information can be introduced into the interaction window with the intelligent agent (such as the interaction window shown by reference numeral 320 in the following example description). An operation related to this at least one piece of historical information (such as pressing Enter) is received, and information about the first intelligent component is displayed.

[0073] As an example, the information of the first intelligent component can be displayed in the following ways: displaying the specific content included in the information of the first intelligent component (such as the name of the first intelligent component, related requirements, execution logic, system prompts (such as prompt words)); displaying a thumbnail of the first intelligent component (such as "Bug Hunter", "ai-bug-fixer skill", etc. in the area shown by label 310 in the following example description). Clicking the thumbnail can display the specific content included in the information of the first intelligent component, such as the name of the first intelligent component, related requirements, execution logic, system prompts, etc.

[0074] The first intelligent component information may correspond to a first intelligent component. This first intelligent component information may be related to a first demand (i.e., the historical demand of the second user). This first demand may be obtained by the intelligent agent based on at least one piece of first historical information.

[0075] According to at least one scenario described herein, receiving a third operation to create a second intelligent component based on at least one first historical information may include: receiving a third operation to create the second intelligent component based on first intelligent component information.

[0076] As can be seen, the method provided in one or more scenarios in this paper can reuse historical intelligent component information to create a second intelligent component, which further reduces the cost of manual description and simplifies the component creation process.

[0077] According to at least one scenario described herein, creating a second intelligent component based on first intelligent component information may include: receiving an operation related to the first intelligent component information and displaying a first configuration item of the first intelligent component; and receiving an operation related to the first configuration item and creating the second intelligent component.

[0078] As an example, as mentioned earlier, the information of the first smart component can be displayed, for example, as a thumbnail. Selecting or clicking the thumbnail will display the first configuration items of the first smart component on the client (i.e., display the configuration items in a form), such as the name of the first smart component, execution logic, system prompts, etc. The first user can adjust the content of the first configuration items in the form, such as modifying or deleting, for example, modifying one or more of the name, execution logic, and system prompts. Finally, receiving this adjustment operation, a second smart component can be created.

[0079] As can be seen, the method provided in one or more scenarios of this paper enables precise manual editing and optimization of the parameters of the newly created component through a form, rather than the existing technical solutions that only support adjustments to the generated content through natural language dialogue. This allows the method provided in one or more scenarios of this paper to bring the following benefits: relying on standardized forms to perform point-to-point editing of component parameters, it is possible to directly locate and modify individual configuration fields, get rid of the rough adjustment mode of pure dialogue-based overall regeneration, and improve the accuracy and controllability of parameter modification.

[0080] According to at least one scenario described herein, receiving an operation related to a first configuration item and creating a second smart component may include: receiving a seventh operation to select at least one third historical information, the at least one third historical information belonging to the one or more historical information; and creating a second smart component based on the first configuration item and at least one third historical information.

[0081] As an example, the seventh operation may include the first user selecting or clicking historical information on the client or a button or control next to that historical information. As another example, the seventh operation may include the first user selecting or clicking historical information on the client or a button or control next to that historical information via voice or other instructions, such as "Please help me select 'ai-bug-fixer' from the official skill".

[0082] Upon receiving the seventh operation, at least one third historical information can be selected; for example, the selection of the third historical information can be indicated by making the third historical information display a different color than the unselected historical information, such as making the selected third historical information gray or black; or, for another example, the selection of the third historical information can be indicated by changing the state of the control next to the third historical information, such as changing the state of the control from "referenced" to "referenced".

[0083] As an example, the third historical information can be directly imported into the form interface of the first configuration item to adjust the content in the first configuration item, thereby creating the second smart component.

[0084] As another example, the third historical information can be introduced into the interactive window next to the first configuration item that interacts with the intelligent agent in natural language (such as the interactive window shown in the example description below, labeled 320) to adjust the content in the first configuration item and thus create a second intelligent component.

[0085] As can be seen, the methods provided in one or more scenarios presented in this paper can reference historical information multiple times at any point in time during form-based interactions, thereby creating a second intelligent component based on this repeatedly referenced historical information. This allows the methods provided in one or more scenarios presented in this paper to bring the following benefits:

[0086] (1) Multiple historical information can be retrieved at any time during the form editing process as a configuration reference without having to exit the form page and restart the creation process, thus optimizing the continuity of component configuration editing.

[0087] (2) Integrate multiple verified historical session data to populate the form configuration content, reduce the workload of manual input by users, and reduce parameter deviation caused by manual input.

[0088] According to at least one scenario described herein, receiving an operation related to the first configuration item and creating a second smart component may include the following steps:

[0089] Step 1: Receive operations related to the first configuration item and display the updated first configuration item.

[0090] As an example, the first user can adjust the content of the first configuration item in the form, such as modifying or deleting it, including changing one or more of the following: name, execution logic, system prompts (such as prompt words). Finally, receiving this adjustment operation will display the updated first configuration item.

[0091] Step 2: Receive the eighth operation and display the third operation information;

[0092] As an example, the eighth operation may include input from the first user (such as manual or voice input), for instance, input in the interactive window shown by reference numeral 320, which will be described later with reference to the example. This third operation information may be displayed during or after the input process. This third operation information may reflect further new needs or intentions from the first user, such as requesting further optimization of system prompts to further regulate behavioral constraints. The aforementioned input operations and the display of the third operation information are conducted through a dialogue with the intelligent agent.

[0093] Step 3: Based on the third operation information and the updated first configuration item, create the second smart component.

[0094] As can be seen, the methods provided in one or more scenarios presented in this paper allow users to first interact with configuration items via a form to modify their content; then, they can further modify the content corresponding to the configuration items via a session; and finally, a second intelligent component is created. This allows the methods provided in one or more scenarios presented in this paper to bring the following benefits:

[0095] (1) It supports seamless operation between two modes: precise fixed-point editing of forms and natural language adjustment of dialogue, breaking the limitation of the two editing modes being separated and unable to be used alternately, thus improving editing flexibility.

[0096] (2) First, the structured parameters are accurately corrected through the form, and then the overall logic and detailed requirements are supplemented and optimized with the help of the dialogue, taking into account both the accuracy of modification and the ability to adjust flexibly, so as to improve the perfection of the final generated intelligent component.

[0097] According to at least one of the scenarios described herein, method 200 may further include the following steps:

[0098] Step 1: Display the information of the second intelligent component, wherein the information of the second intelligent component may correspond to the second intelligent component.

[0099] As an example, the second intelligent component information can be the same as the aforementioned first intelligent component information. As another example, the second intelligent component information can be different from the aforementioned first intelligent component information; in this case, the second intelligent component information is obtained by adjusting the first intelligent component information (such as by modifying one or more of the aforementioned name, execution logic, or system prompt information).

[0100] As an example, the information of the second intelligent component can be displayed in the following ways: displaying the specific content included in the information of the second intelligent component (such as the name of the second intelligent component, related requirements, execution logic, system prompts, etc.); displaying a thumbnail of the second intelligent component, and clicking the thumbnail can display the specific content included in the information of the second intelligent component, such as the name of the second intelligent component, related requirements, execution logic, system prompts, etc.

[0101] Step 2: Receive operations related to the second intelligent component information and display the second configuration item of the second intelligent component. Step 3: Receive operations related to the second configuration item and update the second intelligent component.

[0102] As an example, as mentioned earlier, the information of the second smart component can be displayed, for instance, as a thumbnail. Selecting or clicking the thumbnail displays the second configuration items of the second smart component on the client (i.e., displayed as a form), such as the name of the second smart component, execution logic, system prompts, etc. The first user can adjust the content of the second configuration items in this form, such as modifying or deleting, for example, modifying one or more of the name, execution logic, or system prompts. Finally, receiving this adjustment operation allows the second smart component to be updated.

[0103] As can be seen, the method provided according to one or more scenarios in this paper can update the second intelligent component via the second intelligent component information and the second configuration item. This allows the method provided according to one or more scenarios in this paper to bring the following benefits:

[0104] (1) The overall information of the component and the detailed configuration items are displayed in layers, realizing the step-by-step visualization operation from component overview to single configuration parameter, and the configuration modification path is clear and intuitive.

[0105] (2) Component updates can be completed for independent configuration items without regenerating all component content. This enables fine-grained modification of component parameters, improves editing efficiency and controllability, and further breaks away from the crude adjustment mode of pure dialog-based overall regeneration.

[0106] According to at least one scenario described herein, the second intelligent component may include a functional component of the intelligent system (such as the aforementioned Skill functional component). This functional component can be used to implement at least one function of the intelligent system (such as an intelligent agent).

[0107] The second configuration item may include the name of the functional component, the trigger description, and the execution logic.

[0108] The name can be used to identify the functional component, such as an identification number (ID); for example, the name may include the aforementioned implementation of the user authentication module, refactoring the API interface, ai-bug-fixer (an official skill), etc.

[0109] This trigger description can be used to determine the conditions under which a functional component is invoked. For example, the trigger description could include: triggering when a user provides a plain text table or a screenshot of text-based table data that needs to be converted to a standard Markdown table format.

[0110] This execution logic can be used to perform the corresponding function when the functional component is invoked. For example, the corresponding functions are: 1. Recognize the header delimiter and data line breaks in the input content; 2. Split the header field and each row's data cell by row; 3. Concatenate the | delimiter to generate a standard Markdown table structure; 4. Automatically align column widths and output formatted table text.

[0111] As can be seen, the methods provided in one or more scenarios of this paper enable the creation of functional components of an intelligent system via configuration items (i.e., forms). During the creation process, the name, trigger description, and execution logic of the functional component can be adjusted. This allows the methods provided in one or more scenarios of this paper to offer the following benefits:

[0112] (1) Standardized form configuration items are used to carry the name, trigger description, and core information of the execution logic of the functional components. It supports point-to-point independent modification of a single parameter without the need to regenerate the entire component content, thus achieving higher editing precision.

[0113] (2) The form editing capability is opened synchronously within the functional component creation process, and the three types of core configuration information can be adjusted in real time without waiting for the component to be generated before jumping to the management page to modify it, thus shortening the component iteration cycle.

[0114] According to at least one of the scenarios described herein, the second intelligent component may include a sub-intelligent system of the intelligent system (such as the aforementioned SubAgent).

[0115] As an example, the sub-intelligent system is able to invoke one or more functional components; the functional components can be used to implement at least one function of the sub-intelligent system.

[0116] The second configuration item may include the name of the sub-intelligent system, system prompts, and tool permissions.

[0117] The name can be used to identify the sub-intelligent system, such as an identification number (ID); for example, the name can include the aforementioned "Bug Hunter".

[0118] The system prompts can be used to define the role positioning and behavioral constraints of the sub-intelligent system. For example, the system prompts could include: You are an internal financial audit specialist, responsible only for verifying employee expense reimbursement documents. Work constraints: 1. Strictly verify the upper limits of transportation, accommodation, and meal expenses according to the company's travel reimbursement standards; 2. When there are missing receipts, amounts exceeding the limit, or inconsistent reasons, list the non-compliant items one by one and inform the employee of the correction requirements; 3. Do not answer HR or procurement-related questions outside of the reimbursement policy; 4. Output results should be clear and organized, distinguishing between compliant and abnormal items, and do not modify reimbursement data yourself.

[0119] This tool permission can be used to limit the scope of tool resources, which is the scope that the sub-intelligent system can call. For example, the tool permission may include: only allowing access to three types of tools: ① expense reimbursement standard query tool, ② invoice optical character recognition (OCR) text recognition tool, ③ department budget query tool; and prohibiting access to customer contract tool, payroll calculation tool, and external email sending tool.

[0120] When the sub-intelligent system is scheduled by the intelligent system, it performs processing for specific tasks based on system prompts and tool permissions. For example, after the main intelligent system (such as the aforementioned intelligent agent) receives a session instruction from the user to upload an invoice and submit a reimbursement application, it schedules its own sub-intelligent system (such as the aforementioned subagent) to complete the reimbursement document recognition, expense compliance verification, budget occupancy judgment, and output the reimbursement review result.

[0121] As can be seen, the methods provided in one or more scenarios of this paper enable the creation of sub-intelligent systems within an intelligent system via configuration options (i.e., forms). During the creation process, the name, system prompts, and tool permissions of this sub-intelligent system can be adjusted. This allows the methods provided in one or more scenarios of this paper to offer the following benefits:

[0122] (1) The sub-intelligent system name, system prompt information, and tool permission core parameters are carried by standardized form configuration items. It supports individual location modification of any configuration item without the need to regenerate the entire sub-intelligent system content, which significantly improves the accuracy and controllability of editing.

[0123] (2) Provide real-time form editing capabilities in the sub-intelligent system creation process, allowing for adjustments to role positioning, behavioral constraints, and tool call scope at any time. This eliminates the need to jump to the management page for secondary modifications after creation, simplifying the iteration process and shortening development time.

[0124] According to at least one scenario described herein, method 200 may further include: receiving a fourth operation and displaying first operation information. In this case, the aforementioned receiving of a third operation, based at least on at least one piece of first historical information, to create the second intelligent component, may include: receiving a third operation, based at least on at least one piece of first historical information and the first operation information, to create the second intelligent component.

[0125] As an example, the fourth operation may include input from the first user (such as manual or voice input), for instance, input in the interactive window shown by reference numeral 320, which will be described later with reference to the example. This first operation information may be displayed during or after the input; this first operation information may reflect the operation the first user wants to perform based on the first historical information (i.e., the user's new needs or intentions). This first operation information may include: "Based on this session (i.e., historical information), create a skill for me." Both the input operation and the display of the first operation information are performed through dialogue with the intelligent agent.

[0126] As can be seen, the methods presented in one or more scenarios of this paper can adjust historical information in a conversational manner based on new user needs or intentions to create a second intelligent component. This allows the methods presented in one or more scenarios of this paper to bring the following benefits:

[0127] (1) Based on natural language dialogue, existing historical conversation information can be supplemented and corrected without requiring users to reiterate the task process completely, thus reducing the input cost of creating new intelligent components.

[0128] (2) The second intelligent component is generated by combining new requirements with historical valid data, making full use of existing verification processes and effectively reducing the scene adaptation deviation caused by manual description.

[0129] According to at least one of the scenarios described herein, method 200 may further include the following steps:

[0130] Step 1: Receive the fifth operation and select at least one second historical information, which belongs to one or more historical information.

[0131] The fifth action may include the first user selecting or clicking historical information on the client or a button or control next to that historical information. As another example, the fifth action may include the first user selecting or clicking historical information on the client or a button or control next to that historical information via voice or other instructions, such as "Please help me select 'Refactor API Interface'".

[0132] Upon receiving the fifth operation, at least one second historical information can be selected; for example, the selection of the second historical information can be indicated by making the second historical information display a different color than the unselected historical information, such as making the selected second historical information gray or black; or, for another example, the selection of the second historical information can be indicated by changing the state of the control next to the second historical information, such as changing the state of the control from "referenced" to "referenced".

[0133] Step 2: Receive the sixth operation and display the second operation information.

[0134] As an example, the sixth operation may include the first user's input (such as manual or voice), for instance, input in the interactive window shown by reference numeral 320, which will be described in conjunction with the example later. This second operation information may be displayed during or after the input; this second operation information may reflect the first user's desired action based on the second historical information (i.e., the user's supplementary new needs or intentions). This second operation information may include: "Based on this session (i.e., historical information), please optimize and adjust the previously adjusted skill again." Both the input operation and the display of the second operation information are performed through dialogue with the intelligent agent.

[0135] In the above context, the aforementioned receiving of a third operation, based on at least one first historical information and a first operation information, to create a second intelligent component, may include: receiving a third operation and creating a second intelligent component based on at least one first historical information, at least one second historical information, the first operation information, and the second operation information.

[0136] As an example, receiving a third operation to create a second intelligent component based on at least one first historical information, at least one second historical information, first operation information, and second operation information may include: obtaining an intermediate intelligent component based on at least one first historical information and first operation information; and creating a second intelligent component based on the intermediate intelligent component, at least one second historical information, and second operation information.

[0137] As can be seen, the methods provided in one or more scenarios of this paper can reference historical information multiple times at any point in time in a dialogical manner, thereby creating a second intelligent component based on this repeatedly referenced historical information. This allows the methods provided in one or more scenarios of this paper to bring the following benefits:

[0138] (1) During the entire process of component creation, multiple historical session information can be retrieved at any time as the basis for generation. Reference materials can be supplemented without restarting the creation task, which improves the completeness of the generated content of the component.

[0139] (2) Multiple sets of historical information that have been verified in practice are combined to generate a second intelligent component, making full use of multiple sets of mature task logic, and effectively improving the adaptability of the component to complex business scenarios.

[0140] For ease of understanding, the above methods provided in at least one scenario of this paper are illustrated below with reference to figures.

[0141] Based on the above combination Figure 2 As described, this paper proposes an interactive component generation method based on historical information (such as historical sessions) for one or more scenarios. The system collaborates around a "component configuration state" (JSON object) organized according to a predefined schema, which serves as the data source for all read and write operations during the creation process. The solution provided in one or more scenarios of this paper can include four phases: a session context extraction phase, a requirements parsing and generation phase, a dialogue correction phase, and a confirmation and saving phase.

[0142] Figures 3 to 5 The illustrations provide a schematic diagram related to the session context extraction phase for at least one scenario described herein. Figures 6 to 7 The illustrations provide schematic diagrams related to the requirements analysis and generation phases for at least one scenario presented in this paper. Figures 8 to 9 The illustrations provide a schematic diagram related to the dialogue revision phase for at least one scenario presented in this paper. Figures 10 to 11 The illustrations provide a schematic diagram related to the confirmation and preservation phase for at least one scenario described herein.

[0143] Reference Figures 3 to 10 The leftmost control contains "Sub agents" and "Skills" controls. Clicking the "Sub agents" control allows you to create a sub-intelligent system component, while clicking the "Skills" control allows you to create a functional component.

[0144] Figure 3 and Figure 4 The image on the right shows the smart creation function component displayed after clicking the "Skills" control. This image includes three areas: areas 310, 320, and 330. Area 310 displays the interaction content with the smart agent; area 320 displays the referenced historical sessions and user operation information; and area 330 displays historical information (including historical sessions and official skills), configuration items in the form, etc. Figure 3 and Figure 4 The difference is that, Figure 3 Area 330 shows multiple historical sessions; while Figure 4 Area 330 shows several official skills.

[0145] like Figure 3 As shown in area 330, historical sessions and official skills can be triggered from different locations. Alternatively, both historical sessions and official skills can be triggered from the same location to display simultaneously.

[0146] like Figure 3 As shown in area 330, multiple historical sessions are currently displayed. Displaying these multiple historical sessions could be triggered by a receiving user. Figure 3 The control in the top right corner (as described above) Figure 2 The first operation shown).

[0147] As indicated by the gesture arrow in area 330, the current user clicked on the "Optimize Database Query Tool" historical session (as described above). Figure 2 (The second operation shown). After selecting this history session, the control next to it will change from "referenced" to "referenced".

[0148] Figure 5 The right side shows a diagram for intelligently creating sub-agent systems after clicking the "Sub agents" control. At this time, area 330 only displays historical sessions. Optionally, depending on the actual scenario, both historical sessions and official subagents can be displayed in area 330; historical sessions and official subagents can be triggered from different locations. Alternatively, both historical sessions and official subagents can be triggered and displayed simultaneously from the same location. Figure 5 The remaining displayed content is the same as the above regarding Figure 3 The content described is consistent, so I will not repeat it here.

[0149] In the Figures 3 to 5 The diagram shown relates to the session context extraction phase. When a user references a historical session, the system extracts the task intent (user message, user request, or user intent, i.e., the user's historical request or intent information) and optional execution processes (such as tool call logs, including tool name, input parameter mode, return results, etc.) from the session message stream.

[0150] In cases with long historical sessions (e.g., session length exceeding a predetermined threshold length, which is related to the intelligent agent), summary compression can be performed to organize the extracted results into a structured context, including task objectives, a tool list (such as the Model Context Protocol (MCP) used, the skills invoked, etc.), and success or failure signals (such as failure signals due to errors, data sensitivity, network latency, etc.). Furthermore, the content of the session can be kept within the context window of the intelligent agent model; the intelligent agent model has limited memory, and the length of the content needs to be controlled within the number of characters the model can hold to avoid unnecessary errors.

[0151] Referencing historical sessions and extracting their context can be triggered when creating a smart component and in any subsequent modification round. The extracted historical session context is then appended as incremental context to the current round's generation input to further optimize and adjust the smart component.

[0152] refer to Figure 6 and Figure 7 After referencing historical sessions, the intelligent agent can invoke a model (such as an LLM model) to parse the intent from the historical session and identify the configuration items to be generated. The system predefines configuration schemas (JSON Schema) for both Skill and SubAgent, declaring the name, type, and constraints of each field (e.g., the name of the SubAgent, as described above). Figure 2 The described name), description (optional), system_prompt (as described above) Figure 2 The system prompts described), tools[] (as described above) Figure 2 (The described tool permissions).

[0153] refer to Figure 6 In region 310, the referenced historical session (i.e., the optimized database query function) and a thumbnail of the Bug Hunter generated by the intelligent agent based on the session content are displayed (as described above). Figure 2 The information of the first intelligent component (as stated in the abbreviation). Figure 1 The configuration items are displayed (either simultaneously or after clicking the thumbnail), as shown in the form of area 330. Area 330 shows the configuration items for the Subagent system (as described above). Figure 2The second configuration item includes a name (i.e., Bug Hunter), an optional description, system prompts (Agent prompt), MCP, Skills, etc. The "description" is used to externally display the overall positioning, applicable business scenarios, and core capabilities of the sub-intelligent system, allowing users to quickly identify and differentiate between different sub-intelligent systems. It serves only as descriptive text, does not participate in model inference, and does not restrict tool access permissions. It is mainly used in client lists, selection pop-ups, and form creation to facilitate users' quick decision on whether to select the sub-intelligent system. For example, for a contract compliance review sub-intelligent system, the "description" could include: specifically designed for identifying risks in procurement and service contracts and checking for loopholes in rights and responsibilities; it only processes standard enterprise contract texts and is not responsible for contract amount calculation or stamping processes. (Continue to refer to...) Figure 6 In area 330, a "Create Subagent" control is displayed on the right. If the user confirms that the adjusted smart component is available, they can directly click this control to create the smart component (as described above). Figure 2 As shown, receive the third operation and create the second smart component.

[0154] refer to Figure 7 In zone 310, the referenced historical session (i.e., optimizing database query functionality) and user request (based on this session, help me create a skill) are displayed (as described above). Figure 2 The first operation information and the intelligent component information generated by the intelligent agent based on the session content, the displayed intelligent component information may include a summary of issues obtained based on the analysis of historical sessions (such as...). Figure 7 The document highlights issues such as missing composite indexes, N+1 query problems, and excessive data volume, which will be addressed for future optimization. Figure 7 Use " (This refers to the specific content), and a thumbnail about the AI-bug-fixer skill. This can be compared with the thumbnail. Figure 1 The configuration items are displayed (either simultaneously or after clicking the thumbnail), as shown in the form of area 330. Area 330 shows the configuration items for the functional component `ai-bug-fixer skill` (as described above). Figure 2The second configuration item includes a name (i.e., the ai-bug-fixer skill), an optional description, and a trigger description (e.g., auto trigger). The "description" is a general description of the functional component, used to present the applicable scenarios and functional boundaries on the client interface, allowing users to quickly distinguish between different functions. It does not participate in large model inference or functional operation control. For example, for the ai-bug-fixer skill, the "description" could include: all network operations must be processed through this skill, including: searching, web page crawling, post-login operations, and network interactions. Trigger scenarios: users requesting to search for information, view web page content, access websites requiring login, operate web page interfaces, crawl social media content, read dynamically rendered pages, and any network task requiring a real browser environment. As another example, for the PDF document full-text extraction skill, the "description" could include: extracting all text content from an uploaded PDF file, supporting OCR text recognition for image-text PDFs, processing only local PDF files, and not supporting encrypted or corrupted document parsing.

[0155] Continue to refer to Figure 7 In area 330, a "Create Skill" control is displayed on the right. If the user confirms that the adjusted smart component is available, they can directly click this control to create the smart component (as described above). Figure 2 As shown, receive the third operation and create the second smart component.

[0156] refer to Figure 8 and Figure 9 It displays diagrams related to dialogue correction. Users can provide suggestions for modification in natural language, and the intelligent agent regenerates based on the results of the previous round, repeating this process several times. During this period, users can again select historical conversations or official skills for reference, and the views on both sides are refreshed in real time.

[0157] On the form side: Configure JSON to be rendered as form controls by mapping each field according to the schema (e.g., in a SubAgent scenario) (as described above). Figure 2 (As mentioned in the second configuration item), when a user edits a field, a status change is submitted.

[0158] On the dialogue side: The user's natural language modification request, along with the latest current state (including the result of manual editing), is sent to the intelligent agent (such as its generation module), and the returned difference is then written back to the state.

[0159] refer to Figure 8Zone 310 shows a previously referenced historical session (i.e., optimized database query functionality). Now, the user wants to reference another historical session; as indicated by the gesture arrow in Zone 330, the current user has clicked on the "Refactor API Tooling" historical session (as described above). Figure 2 (The fifth operation and the second history information are shown). After selecting this history session, the control next to it will change from "referenced" to "referenced", and an indicator that the history session has been referenced will be displayed in area 320, as shown in the figure. After the user enters their further requirements in area 320, operation information can be displayed in area 320 (as described above). Figure 2 As mentioned above, receiving the sixth operation and displaying the second operation information shows that the user can make adjustments on the dialogue side.

[0160] refer to Figure 9 In area 330, the rendered form controls are shown in a form-like manner (as described above). Figure 2 (The second configuration item mentioned above). Users can edit any control to make targeted modifications. Once the modification is complete, a state change is submitted, and the corresponding smart component's configuration item content is updated.

[0161] refer to Figure 10 and Figure 11 The diagram illustrates the process related to the confirmation and saving phase. Once the user confirms that the adjusted smart components are working correctly, they can write the configuration to the definition file and register it with the platform, enabling it to be called and executed normally.

[0162] refer to Figure 10 After clicking the "Create Subagent" control, a pop-up window will appear. This window allows users to set basic information about the created subagent, such as tags (e.g., skill-creator) and visibility scope (e.g., only individuals, teams, organizations, etc.). Once this basic information is set, the component can be registered and added to the smart agent platform.

[0163] Figure 11 This indicates that the smart component subagent has been successfully created. Users can view the details of the smart component and also test the created smart component.

[0164] It should be noted that, Figures 3 to 10 The area indicated by label 320 is the interaction window area where the user interacts with the intelligent agent in natural language; in this area, referenced historical information (such as...) can be displayed. Figure 8 The historical session "Reconstruct API" interface shown), and user requirement information (as described above) Figure 2The described first operation information, second operation information, and third operation information, etc. Click the Enter key in area 320 (i.e., in the image). The displayed information can be transmitted to the intelligent agent for interaction. Furthermore, the arrangement of areas 310, 320, 330, and various controls described above is merely an example. Any adjustments based on actual needs are permitted.

[0165] Figure 12 The flowchart illustrating a portion of the process for creating a smart component is shown in at least one scenario described herein.

[0166] refer to Figure 12 Initially, there are two ways to initiate creation: Step 1 (S1210), initiating creation (Method 1), describing the requirements in natural language; and Step 1 (S1220), initiating creation (Method 2), selecting a historical session and extracting the requirements. Then, in S1230, Step 2, Requirement Analysis and Follow-up Questions, the LLM analyzes the intent and, if insufficient, asks the user for follow-up questions. Next, in S1240, Step 3, Configuration Generation, the LLM generates a complete configuration and can output a preview in the dialog stream. Then, in S1250, Step 4, Dialogue-Based Correction, the user can provide modification suggestions in natural language, and the model can be regenerated based on the previous round's results and modification suggestions. After several rounds of iterations in S1240 and S1250, the user's final requirements are met. Finally, in S1260, Step 5, Confirmation and Saving, the created component is written to the definition file and registered. Subsequent adjustments to the created component require manual editing on the management page.

[0167] Based on the above combination Figures 3 to 12 As can be seen from the examples described, the methods provided in one or more scenarios in this paper can achieve the following:

[0168] (1) Dual-channel synchronous modification mechanism: Dialogue modification and manual form editing operate in the same configuration state, and are synchronized in real time. Changes in either channel become the basis for subsequent operations in the other channel, and can be used interchangeably.

[0169] (2). Dynamic injection of session context at multiple times: Historical sessions can be referenced at the time of creation and at any round of modification. The historical sessions referenced are extracted in a structured manner and used as the basis for incremental generation to correct existing content or create new content.

[0170] (3) Structured generation of schema constraints and field-level backfilling: The LLM output is constrained by the configured schema, and each field is mapped and backfilled to the form. The field level is auditable, verifiable, and editable in real time.

[0171] (4) Differential generation of official template base: The official Skill can be previewed and selected as the initial base. LLM only generates differential modifications and merges them, instead of regenerating the whole.

[0172] (5) Create-Debug-Iteration Closed Loop: After saving, an associated debugging session is automatically established. The debugging session can be used as a new reference context to flow back to the generation process.

[0173] The resulting beneficial effects include at least the following:

[0174] (1) Context can be added at any time, and users can gradually calibrate the generated results with verified real session data, which significantly reduces the number of correction rounds and improves the consistency between configuration and actual scenario.

[0175] (2) Through multiple rounds of dialogue, the two channels each have their own advantages and there is no state conflict, thus improving both creation efficiency and accuracy.

[0176] In addition to the methods described above, this document also provides corresponding equipment, media, and products for one or more scenarios. These will be explained below with reference to the accompanying drawings.

[0177] Figure 13 A schematic block diagram of a device 1300 for creating intelligent components, provided in at least one of the scenarios described herein, is shown. The description of method 200 above also applies to device 1300, unless otherwise explicitly stated.

[0178] like Figure 13 As shown, the device 1300 in this case may include a first operation unit 1310, a second operation unit 1320 and a third operation unit 1330.

[0179] According to at least one of the embodiments described herein, the first operation unit 1310 may be configured to receive a first operation and display one or more historical information.

[0180] According to at least one of the embodiments described herein, the second operation unit 1320 may be configured to receive a second operation to select at least one first historical information, wherein the at least one first historical information belongs to the one or more historical information and is related to the creation of the first intelligent component.

[0181] According to at least one of the scenarios described herein, the third operation unit 1330 can be configured to receive a third operation to create a second intelligent component based at least one first historical information.

[0182] It should be noted that, for clarity and brevity, not all the constituent modules or units of the device 1300 are shown in at least one scenario herein. To achieve the necessary functions of the information device, those skilled in the art may provide or configure other constituent units (not shown) according to specific needs, and this at least one scenario herein does not impose any limitations on this.

[0183] Figure 14 A schematic block diagram of an electronic device 1400 in at least one of the present invention is shown.

[0184] Reference Figure 14 The electronic device 1400 may include a processing device 1401 and a storage device 1408. As needed, the electronic device 1400 may also include a read-only memory (ROM) 1402, a random access memory (RAM) 1403, a bus 1404, an (I / O) interface 1405, an input device 1406, an output device 1407, and a communication device 1409.

[0185] Figure 14 The electronic device shown is merely an example and should not impose any limitations on the functionality and scope of use in the context of this article.

[0186] For example, processing device 1401 may be a central processing unit (CPU), digital signal processor (DSP), image processor (GPU), general-purpose graphics processor (GPGPU), or other form of processing unit with data distribution capability and / or instruction execution capability. It may be a general-purpose processor or a dedicated processor and may control other components in the electronic device to perform the desired function. Electronic device 1400 may include one or more processing devices 1401.

[0187] For example, storage device 1408 may include one or more computer program instructions, which may be stored in various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc. The processing device may execute the one or more computer program instructions to implement the functions described herein (implemented by the processing device) and / or other desired functions, such as methods for creating intelligent components. Various application programs and various data, such as the aforementioned skills and subagents, may also be stored in the computer-readable storage medium; this is not limited in this context.

[0188] In addition, the processing device 1401 can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 1402 or the program loaded from the storage device 1408 into the random access memory (RAM) 1403. The RAM 1403 also stores various programs and data required for the operation of the electronic device 1400. The processing device 1401, ROM 1402, and RAM 1403 are interconnected via a bus 1404. An input / output (I / O) interface 1405 is also connected to the bus 1404.

[0189] The following devices can be connected to I / O interface 1405: input devices 1406 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 1407 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1408 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1409. Communication device 1409 allows electronic device 1400 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 14 An electronic device 1400 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0190] Specifically, according to the context of this document, the process described in the above-referenced flowchart can be implemented as a computer software program. For example, the context includes a computer program product comprising one or more computer instructions that can be used to perform the method shown in the flowchart. In such a scenario, the one or more computer instructions can be downloaded and installed from a network via communication device 1409, or installed from storage device 1408, or installed from ROM 1402. When the one or more computer instructions are executed by processing device 1401, the functions defined in the method of this context are performed.

[0191] At least one aspect of this document also provides a non-transitory computer-readable storage medium. This storage medium stores non-transitory computer-readable instructions. When executed by a processor, the computer-readable instructions can perform the method described with reference to the above figures according to this aspect. The storage medium in this aspect can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. 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. Volatile memory can be random access memory (RAM) 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 dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct memory bus random access memory (DR RAM). It should be noted that the memory of the methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0192] It should be noted that the computer-readable medium described above in this document can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this document, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0193] The aforementioned computer-readable medium carries one or more computer program products, which include one or more computer instructions that, when executed by a processor, perform any of the methods described above.

[0194] Computer program code for performing the operations described herein may be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0195] The flowcharts and block diagrams in the accompanying figures illustrate the architecture, functionality, and operation of possible implementations of the systems, methods, and computer program products according to the various scenarios described herein. In this respect, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the figures. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of block numbers in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0196] The units or modules described in this document can be implemented in software or hardware. The names of the units or modules do not necessarily constitute a limitation on the unit or module itself.

[0197] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0198] In the context of this document, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction 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 be, 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 machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0199] According to one or more scenarios described herein, Example 1 provides a method for creating a smart component, comprising: receiving a first operation to display one or more historical information; receiving a second operation to select at least one first historical information, wherein the at least one first historical information belongs to the one or more historical information and is related to the creation of a first smart component; and receiving a third operation to create a second smart component based at least on at least one first historical information.

[0200] According to one or more scenarios in this article, Example 2 provides that Example 1 further includes: receiving an operation related to at least one first historical information and displaying first smart component information, wherein the first smart component information corresponds to a first smart component, the first smart component information is related to a first requirement, and the first requirement is obtained based on at least one first historical information; wherein receiving a third operation, based at least one first historical information, and creating a second smart component includes: receiving a third operation and creating a second smart component based on the first smart component information.

[0201] According to one or more scenarios in this article, Example 3 provides the creation of a second smart component based on the first smart component information in Example 2, including: receiving an operation related to the first smart component information, displaying a first configuration item of the first smart component; and receiving an operation related to the first configuration item to create a second smart component.

[0202] According to one or more scenarios in this article, Example 4 provides that Example 1 also includes: displaying second smart component information, wherein the second smart component information corresponds to the second smart component; receiving operations related to the second smart component information; displaying the second configuration item of the second smart component; receiving operations related to the second configuration item; and updating the second smart component.

[0203] According to one or more scenarios in this paper, Example 5 provides the second intelligent component in Example 4, which includes a functional component of the intelligent system. The functional component is used to implement at least one function of the intelligent system. The second configuration item includes the name of the functional component, a trigger description, and execution logic. The trigger description is used to determine the conditions under which the functional component is invoked, and the execution logic is used to execute the corresponding function when the functional component is invoked.

[0204] According to one or more scenarios in this document, Example 6 provides a sub-intelligent system of the intelligent system, which is the second intelligent component in Example 4. The sub-intelligent system can call one or more functional components, which are used to implement at least one function of the sub-intelligent system. The second configuration item includes the name of the sub-intelligent system, system prompt information and tool permissions. When the sub-intelligent system is scheduled by the intelligent system, it performs processing for a specific task according to the system prompt information and tool permissions.

[0205] According to one or more scenarios in this article, Example 7 provides that Example 1 also includes: receiving a fourth operation and displaying first operation information; wherein receiving a third operation and creating a second intelligent component based on at least one first historical information includes: receiving a third operation and creating a second intelligent component based on at least one first historical information and first operation information.

[0206] According to one or more scenarios in this document, Example 8 provides that Example 7 further includes: receiving a fifth operation to select at least one second historical information, the at least one second historical information belonging to one or more historical information; and receiving a sixth operation to display second operation information; wherein receiving a third operation to create a second smart component based at least one first historical information and first operation information includes: receiving the third operation to create a second smart component based on at least one first historical information, at least one second historical information, first operation information and second operation information.

[0207] According to one or more scenarios in this document, Example 9 provides the operation of receiving an operation related to a first configuration item in Example 3 to create a second smart component, including: receiving a seventh operation to select at least one third historical information, the at least one third historical information belonging to the one or more historical information; and creating a second smart component based on the first configuration item and at least one third historical information.

[0208] According to one or more scenarios in this document, Example 10 provides the method of receiving an operation related to the first configuration item and creating a second smart component as in Example 3, including: receiving an operation related to the first configuration item and displaying the updated first configuration item; receiving an eighth operation and displaying third operation information; and creating the second smart component based on the third operation information and the updated first configuration item.

[0209] According to one or more scenarios described herein, Example 11 provides a device for creating a smart component, comprising: a first operation unit configured to receive a first operation to display one or more historical information; a second operation unit configured to receive a second operation to select at least one first historical information, wherein the at least one first historical information belongs to the one or more historical information and is related to the creation of a first smart component; and a third operation unit configured to receive a third operation to create a second smart component based on at least one first historical information.

[0210] According to one or more of the embodiments herein, Example Twelve provides an electronic device including: at least one processor, and a memory storing one or more computer program instructions; wherein the one or more computer program instructions are executed by the at least one processor to perform the method provided in at least one of the embodiments herein.

[0211] According to one or more of the scenarios described herein, Example Thirteen provides a non-transitory computer-readable storage medium that non-transitory stores computer-readable instructions, wherein the methods provided in at least one scenario described herein are implemented when the computer-readable instructions are executed by a processor.

[0212] According to one or more scenarios herein, Example Fourteen provides a computer program product comprising one or more computer instructions, wherein the one or more computer instructions are executed by a processor to perform the method provided in at least one scenario herein.

[0213] The above description is merely a preferred embodiment and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure herein is not limited to technical solutions formed by specific combinations of the above-described technical features, but also includes other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed herein that have similar functions.

[0214] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain contexts, multitasking and parallel processing may be advantageous. Similarly, while some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of this paper. Certain features described in the context of a single case can also be implemented in combination within that single case. Conversely, various features described in the context of a single case can also be implemented individually or in any suitable sub-combination in multiple cases.

[0215] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A method for creating smart components, comprising: Upon receiving the first operation, display one or more historical information items; Receive a second operation to select at least one first historical information, wherein the at least one first historical information belongs to the one or more historical information and is related to the creation of the first intelligent component; as well as Receive a third operation to create a second intelligent component based at least on the at least one first historical information.

2. The method according to claim 1, further comprising: Receive an operation related to the at least one first historical information and display first smart component information, wherein the first smart component information corresponds to the first smart component, the first smart component information is related to a first requirement, and the first requirement is obtained based on the at least one first historical information; The third operation, which involves creating a second intelligent component based at least on the at least one first historical information, includes: Upon receiving the third operation, the second intelligent component is created based on the information of the first intelligent component.

3. The method according to claim 2, wherein, The step of creating the second intelligent component based on the first intelligent component information includes: Receive operations related to the information of the first intelligent component, and display the first configuration item of the first intelligent component; and Receive the operation related to the first configuration item and create the second smart component.

4. The method according to claim 1, further comprising: Displays information about the second intelligent component, wherein the information about the second intelligent component corresponds to the second intelligent component. Receive operations related to the information of the second intelligent component and display the second configuration item of the second intelligent component; Receive operations related to the second configuration item and update the second smart component.

5. The method according to claim 4, wherein, The second intelligent component includes functional components of the intelligent system, which are used to implement at least one function of the intelligent system. The second configuration item includes the name of the functional component, a trigger description, and execution logic. The trigger description is used to determine the conditions under which the functional component is invoked, and the execution logic is used to execute the corresponding function when the functional component is invoked.

6. The method according to claim 4, wherein, The second intelligent component includes a sub-intelligent system of the intelligent system, which can call one or more functional components to implement at least one function of the sub-intelligent system; The second configuration item includes the name of the sub-intelligent system, system prompts, and tool permissions. When the sub-intelligent system is scheduled by the intelligent system, it performs processing for specific tasks based on the system prompts and tool permissions.

7. The method according to claim 1, further comprising: Receive the fourth operation and display the information from the first operation. The third operation, which involves creating the second intelligent component based at least on the at least one first historical information, includes: Upon receiving the third operation, the second intelligent component is created based at least on the at least one first historical information and the first operation information.

8. The method according to claim 7, further comprising: Receive the fifth operation and select at least one second historical information, wherein the at least one second historical information belongs to the one or more historical information; as well as Receive the sixth operation and display the second operation information; The process of receiving the third operation and creating the second intelligent component based at least on the at least one first historical information and the first operation information includes: Upon receiving the third operation, the second intelligent component is created based on the at least one first historical information, the at least one second historical information, the first operation information, and the second operation information.

9. The method according to claim 3, wherein, The step of receiving operations related to the first configuration item and creating the second smart component includes: Receive the seventh operation, select at least one third historical information, wherein the at least one third historical information belongs to the one or more historical information; and The second intelligent component is created based on the first configuration item and the at least one third historical information.

10. The method according to claim 3, wherein, The step of receiving operations related to the first configuration item and creating the second smart component includes: Receive operations related to the first configuration item and display the updated first configuration item; Receive the eighth operation and display the third operation information; and Based on the third operation information and the updated first configuration item, the second smart component is created.

11. A device for creating intelligent components, comprising: The first operation unit is configured to receive a first operation and display one or more historical information. The second operation unit is configured to receive a second operation and select at least one first historical information, wherein the at least one first historical information belongs to the one or more historical information and is related to the creation of the first intelligent component; as well as The third operation unit is configured to receive a third operation and create a second intelligent component based at least on the at least one first historical information.

12. An electronic device, comprising: At least one processor, and Memory, which stores one or more computer program instructions; The one or more computer program instructions are executed by the at least one processor to perform the method of any one of claims 1 to 10.

13. A non-transitory computer-readable storage medium for non-transitory storage of computer-readable instructions, wherein, The method of any one of claims 1 to 10 is implemented when the computer-readable instructions are executed by a processor.

14. A computer program product comprising one or more computer instructions, wherein, The one or more computer instructions are executed by the processor to perform the method of any one of claims 1 to 10.

Images