System access method and apparatus, electronic device, storage medium, and program product

Abstract

The application provides a system access method and device, electronic equipment, a storage medium and a program product. The method comprises the following steps: receiving an access request submitted by an access terminal through a front-end page; performing intention recognition on the access request, extracting key information, and configuring parameters in the key information into a system parameter table through a parameter intelligent agent; after the parameter configuration is completed, initiating a joint debugging request message to the access terminal; receiving feedback information of the access terminal on the joint debugging confirmation; wherein the feedback information comprises joint debugging message content; when the feedback information is a confirmation access, sending the joint debugging message content to a joint debugging system by a sending intelligent agent; receiving feedback information returned by the joint debugging system, and judging whether the feedback information is normal; when the feedback information is normal, confirming that the system access is successful. By combining natural language intention recognition, parameter intelligent agent automatic configuration and sending intelligent agent cooperative scheduling, the processing efficiency and system scalability are improved, and the overall cycle of system access is shortened.

Description

Technical Field

[0001] This application relates to the field of system automated configuration, and in particular to a system access method, device, electronic device, storage medium and program product. Background Technology

[0002] In the fields of fintech and payment clearing, system integration and joint testing between internal application systems and external partners are crucial prerequisites for ensuring rapid business launch and stable operation. Taking the general gateway of a payment platform as an example, as the core hub for data interaction between internal and external systems, the integration process typically involves multiple complex technical aspects, such as interface permission allocation, IP whitelist management, digital certificate management, encryption and decryption algorithm integration, message assembly and signing, and initiation of joint testing requests.

[0003] In traditional access scenarios, service providers need to manually operate across multiple heterogeneous management consoles and integrate officially provided software development kits (SDKs) by writing code to complete message structure conversion, sensitive information encryption, and communication protocol adaptation. This process is highly dependent on manual configuration and expert supervision, with each operational step being fragmented and lacking a unified automated collaboration and intelligent verification mechanism. With the surge in the number of access providers and the accelerated pace of agile business iterations, the traditional manually driven access model has gradually exposed its drawbacks, such as lengthy processes, low fault tolerance, and inconsistent toolchains, severely restricting the efficiency of joint debugging and system stability. Summary of the Invention

[0004] This application provides a system access method, device, electronic device, storage medium, and program product to achieve automated, intelligent, and highly efficient access.

[0005] In a first aspect, embodiments of this application provide a system access method, including:

[0006] Receive access requests submitted by access terminals through the front-end page;

[0007] The system identifies the intent of access requests, extracts key information, and configures the parameters in the key information into the system parameter table through a parameter agent. After the parameter configuration is completed, a joint debugging request message is sent to the access terminal.

[0008] Receive feedback information from the access terminal confirming the joint debugging; the feedback information includes the content of the joint debugging message; when the feedback information is confirmation of access, the sending agent sends the content of the joint debugging message to the joint debugging system;

[0009] Receive feedback information returned by the joint debugging system and determine whether the feedback information is normal; if the feedback information is normal, determine that the system access is successful.

[0010] In one embodiment, intent recognition is performed on the access request, and key information is extracted, specifically including:

[0011] The large language model is invoked in conjunction with retrieval enhancement technology to perform semantic parsing on access requests, generating independently executable task decomposition sequences; the task decomposition sequence includes at least one specific configuration task from permission configuration, certificate configuration, and IP whitelist configuration;

[0012] Based on the configuration task, the access terminal identifier, access terminal address, and access system identification information are extracted from the access request.

[0013] In one embodiment, after invoking a large language model to perform semantic parsing on the access request and generating an independently executable task breakdown sequence, the method further includes:

[0014] Based on a pre-built agent registry and vector matching algorithm, the capability requirement features of each subtask in the task decomposition sequence are extracted, and the expert agent node that executes the corresponding subtask is dynamically matched.

[0015] The matched expert agent node addresses and their corresponding execution parameters are bound together to generate structured task routing instructions, which are then sent to the corresponding expert agent nodes for execution.

[0016] In one embodiment, after the parameter configuration is completed, the following is also included:

[0017] Check the completeness of the parameters in the system parameter table;

[0018] When a missing parameter is detected, a parameter supplementation prompt is returned to the user, and the user's supplemented parameter information is received until the parameter is complete.

[0019] In one embodiment, when a missing parameter is detected, a parameter supplementation prompt is returned to the user, and the user's supplemented parameter information is received until the parameter is complete. Specifically, this includes:

[0020] Using the access terminal identifier as an index, query the configuration status of permission configuration items, certificate configuration items, and IP whitelist configuration items in the system parameter table to determine whether the access information of the access terminal is fully configured.

[0021] When any of the permission configuration items, certificate configuration items, or IP whitelist configuration items is detected to be missing or incomplete, the parameters are determined to be incomplete, and a parameter supplementation prompt message is sent to the access terminal. The parameter supplementation prompt message includes: a list of missing configuration item types, the configuration content requirements and format specifications corresponding to each missing item.

[0022] In one embodiment, the method further includes:

[0023] The system receives parameter information submitted by the user to supplement missing configuration items, and the parameter agent performs format verification and validity validation on the supplemented parameters.

[0024] After successful verification, the parameters are updated to the system parameter table and the parameter completeness check is triggered again.

[0025] In one embodiment, complete parameters include: the permission configuration, certificate configuration, and IP whitelist configuration retrieved using the access terminal identifier as an index are all in a configured and verified state.

[0026] In one embodiment, initiating a joint debugging request message to the access terminal specifically includes:

[0027] The intelligent agent sends a joint debugging request message to the access terminal.

[0028] By sending a joint debugging request message to the access terminal through the intelligent agent, specifically including:

[0029] The sending agent converts the request message into a structured message according to the interface specification document; identifies and encrypts sensitive information in the message body, digitally signs the message, assembles the message header and message body according to the interface specification, and generates the joint debugging request message to be sent.

[0030] In one embodiment, the feedback information received from the access terminal confirming the joint debugging includes:

[0031] Receive feedback information returned by the access terminal through a general gateway;

[0032] Parse the response code and sensitive information in the feedback message;

[0033] When the response code is normal, the sensitive information is decrypted and the decrypted response is sent back to the collaborative scheduling agent.

[0034] When the response code is an error code, extract the request keywords and trigger the request diagnostic process.

[0035] In one embodiment, triggering a diagnostic request process specifically includes:

[0036] The diagnostic agent is invoked, and the diagnostic agent receives the request keyword and request time.

[0037] Query the joint debugging system processing logs based on the request keywords and request time, and extract key information such as request messages and exception logs from the processing logs.

[0038] The key information is compared and matched with the error messages and solution caches built into the diagnostic agent to obtain the corresponding solution and display it on the front end.

[0039] In one embodiment, the method further includes:

[0040] When the key information does not match the error message and solution cache, the diagnostic agent returns the default support information to the front end and provides an entry point for manual support.

[0041] Secondly, embodiments of this application provide a system access device, including:

[0042] The receiving module is used to receive access requests submitted by the access terminal through the front-end page;

[0043] The configuration module is used to identify the intent of the access request, extract key information, and configure the parameters in the key information into the system parameter table through the parameter agent; after the parameter configuration is completed, a joint debugging request message is sent to the access terminal.

[0044] The sending module is used to receive feedback information from the access terminal confirming the joint debugging; the feedback information includes the content of the joint debugging message; when the feedback information is confirmation of access, the sending agent sends the content of the joint debugging message to the joint debugging system.

[0045] The processing module is used to receive feedback information returned by the joint debugging system and determine whether the feedback information is normal; when the feedback information is normal, the system connection is confirmed to be successful.

[0046] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0047] The memory stores instructions that the computer executes;

[0048] The processor executes computer execution instructions stored in memory, causing the processor to perform any of the methods described above.

[0049] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement any of the methods described above.

[0050] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements any of the methods described above.

[0051] The system access method, apparatus, electronic device, storage medium, and program product provided in this application embodiment include: receiving an access request submitted by an access terminal through a front-end page; performing intent recognition on the access request, extracting key information, and configuring the parameters in the key information into a system parameter table through a parameter agent; after the parameter configuration is completed, initiating a joint debugging request message to the access terminal; receiving feedback information from the access terminal confirming the joint debugging; wherein the feedback information includes the content of the joint debugging message; when the feedback information is a confirmation of access, the sending agent sends the content of the joint debugging message to the joint debugging system; receiving feedback information returned by the joint debugging system and determining whether the feedback information is normal; when the feedback information is normal, confirming successful system access. By combining natural language intent recognition, automatic parameter agent configuration, and collaborative scheduling of sending agents, the traditional system access process, which relies on manual, step-by-step data entry and is prone to errors, is transformed into a low-threshold, highly accurate, and fully closed-loop automated intelligent access system. Users only need to describe their access requirements in natural language through the front-end page, and the system can automatically extract key information and complete parameter configuration, avoiding format errors and field omissions caused by manual data entry. After parameter configuration, the system proactively initiates a joint debugging request and guides the user to confirm the joint debugging message, forming a seamless closed-loop process from configuration to joint debugging. The user previews and confirms the joint debugging message before it is sent, serving as a pre-verification and exposing potential problems in advance. Each agent collaborates according to its function, breaking down the complex access process into specialized sub-tasks, improving processing efficiency and system scalability. Finally, the system automatically judges the joint debugging feedback and confirms successful access, eliminating the need for manual verification and significantly shortening the overall system access cycle. This improves user experience while effectively reducing maintenance costs. Attached Figure Description

[0052] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0053] Figure 1 A flowchart illustrating a system access method provided in an embodiment of this application;

[0054] Figure 2 An architecture diagram of the system access scheme provided in the embodiments of this application;

[0055] Figure 3 A schematic diagram of the system access device provided in this application;

[0056] Figure 4 A schematic diagram of the structure of the electronic device provided in this application.

[0057] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0058] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application.

[0059] In existing financial system gateway access scenarios, the network architecture typically exhibits a fragmented modular nature, with parameter management, API calls, and log monitoring scattered across independent consoles. Access providers must manually enter permission information across platforms, rely on official SDKs to write encryption and decryption logic, and manually retrieve massive amounts of logs to compare error codes when integration testing fails. This architecture creates a significant semantic gap between business intent and machine instructions. Manual configuration is prone to overlooking key fields or certificate mismatches. SDK dependency forces access providers to possess high coding skills, while troubleshooting relies entirely on expert experience, lacking automated state transitions and contextual memory. This results in frequent integration testing link interruptions, low collaboration efficiency, and difficulty in supporting large-scale open access requirements.

[0060] This application is primarily applied to system integration scenarios within fintech payment and settlement platforms and open banking ecosystems. The network architecture typically includes a front-end interactive terminal, an intelligent scheduling engine deployed in a service middleware, several distributed expert agent nodes, a vector knowledge base and business database, and an underlying general gateway cluster and log center. The front-end terminal is responsible for collecting natural language commands and interactive feedback from the access party; the intelligent scheduling engine, as the core hub, maintains a global context cache and agent registry; expert agent nodes load and execute configuration, sending, or diagnostic tasks as needed; the underlying gateway and log center provide message routing, protocol parsing, and historical operational data support. The overall architecture supports cloud or private deployment, meeting the requirements of high-concurrency, multi-tenant joint debugging and testing environments.

[0061] The system access method provided in this application addresses the pain points of existing access processes, such as fragmentation, high reliance on manual intervention, and low fault tolerance. Recognizing that traditional linear, rule-driven technical paths cannot adapt to complex and ever-changing business interaction scenarios, the inventors propose introducing a large language model as a cognitive center. This abandons the single-script or fixed SDK calling mode and instead designs a multi-agent collaborative architecture. By constructing a dynamic task planning and routing mechanism, discrete steps such as parameter assembly, message construction, and fault diagnosis are decoupled into independent, pluggable dedicated agents. Context caching and feedback verification mechanisms enable seamless task state transitions. During the exploration process, interface specifications, historical logs, and knowledge bases are further vectorized and embedded into the agent workflow, enabling the system to possess semantic understanding, adaptive conversion, and pattern matching capabilities. Ultimately, this evolves into an automated access technology solution driven by natural language and possessing self-verification and self-diagnosis capabilities.

[0062] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0063] like Figure 1 As shown, Figure 1 This is a flowchart illustrating a system access method provided in an embodiment of this application. The system access method is described using a collaborative scheduling agent as the executing entity, and includes the following steps:

[0064] Step S101: Receive the access request submitted by the access terminal through the front-end page.

[0065] Specifically, access terminals (such as merchant systems, third-party platforms, etc.) fill out and submit access applications through a front-end page. The front-end page is the entry point for the entire access process, responsible for collecting basic information and access requests from the access party. The access terminal initiates access through the page, providing information such as appId, access interface name, and access IP address. Access example: "I want to access and debug the emop.unsign.request interface, appId is T0000001, access IP is 172.23.45.67." After the page sends this information to the collaborative agent, the collaborative scheduling agent performs intent recognition and task decomposition to begin parameter configuration.

[0066] Step S102: Perform intent recognition on the access request, extract key information, and configure the parameters in the key information into the system parameter table through the parameter agent; after the parameter configuration is completed, send a joint debugging request message to the access terminal.

[0067] Specifically, the system performs intent recognition on access requests (usually based on NLP technology), automatically extracting key information (such as access type, interface version, IP whitelist, encryption method, etc.) from unstructured or semi-structured request content. Subsequently, a parameter agent automatically writes these parameters into the system parameter table, completing the system-side configuration. After configuration, the system automatically generates and sends a joint debugging request message to the access terminal. Compared to the traditional method where maintenance personnel need to manually interpret the application content and configure parameters one by one, this application automates the process through intent recognition and intelligent agents. It reduces human error: machine extraction and parameter writing avoid omissions and mismatches that occur during manual input; and it shortens the access cycle: the connection from "application → configuration → initiating joint debugging" is completed automatically by the system, eliminating the need for manual scheduling.

[0068] Step S103: Receive feedback information from the access terminal confirming the joint debugging; wherein, the feedback information includes the content of the joint debugging message; when the feedback information is confirmation of access, the sending agent sends the content of the joint debugging message to the joint debugging system.

[0069] Specifically, after receiving the joint debugging request message, the access terminal performs local verification and returns feedback information. This feedback information includes the joint debugging message content (i.e., the test message prepared by the access party). When the feedback information indicates that the access party confirms access is possible, the sending agent automatically forwards the joint debugging message content to the joint debugging system (such as a sandbox environment, test gateway, etc.). This achieves unattended joint debugging: after the access party confirms, there is no need for manual uploading or forwarding of messages; the sending agent automatically completes the delivery. It ensures message integrity and consistency: directly transmitting the original joint debugging message returned by the access party avoids truncation or tampering caused by manual copying and pasting. It creates a closed-loop process: automatically linking the "access party confirmation" and "message sending to the joint debugging system" actions eliminates process breakpoints.

[0070] The content of the integration test message can also be semi-structured tabular data from the interface specification, as shown in the table below:

[0071]

[0072] Step S104: Receive feedback information returned by the joint debugging system and determine whether the feedback information is normal; when the feedback information is normal, confirm that the system access is successful.

[0073] Specifically, after the integration testing system (test environment) executes the integration testing message, it returns a result. The system receives this feedback information and performs automated judgment (such as verifying the return code, response time, and business logic correctness). If the feedback information is normal, it automatically confirms that the system has successfully connected, completing the entire connection process. Automated acceptance: Replaces the traditional manual verification of integration testing results, improving acceptance efficiency. Instant feedback: The connecting party can know the connection result in real time without waiting for manual notification. Quality control: Through preset judgment rules, it ensures that only systems that have fully passed integration testing can be determined as successfully connected, ensuring the stability of the production environment.

[0074] This application combines natural language intent recognition, automatic parameter agent configuration, and collaborative scheduling of sending agents to transform the traditional system access process, which relies on manual, step-by-step data entry and is prone to errors, into a low-threshold, highly accurate, and fully closed-loop automated intelligent access system. Users only need to describe their access requirements in natural language through a front-end page, and the system automatically extracts key information and completes parameter configuration, avoiding formatting errors and field omissions caused by manual data entry. After parameter configuration, the system proactively initiates a joint debugging request and guides the user to confirm the joint debugging message, forming a seamless closed-loop process from configuration to joint debugging. The user previews and confirms the joint debugging message before it is sent, serving as a pre-verification and exposing potential problems in advance. Each agent collaborates according to its function, breaking down the complex access process into specialized sub-tasks, improving processing efficiency and system scalability. Finally, the system automatically judges the joint debugging feedback and confirms successful access, eliminating the need for manual verification and significantly shortening the overall system access cycle. This improves user experience while effectively reducing maintenance costs.

[0075] like Figure 2 As shown, Figure 2This is an architecture diagram of the system access scheme provided in this application embodiment. The system access scheme includes a front-end page, a collaborative scheduling agent, a parameter agent, a sending agent, a diagnostic agent, and a large model. The front-end page is mainly used for interaction with the user, receiving user access requests and integration information through a question-and-answer dialogue, and providing feedback on the access scheme and integration results to the user. After receiving a user message, the front-end page sends it to the collaborative scheduling agent and receives the response from the collaborative scheduling agent. The collaborative scheduling agent is responsible for understanding user intent, planning tasks, scheduling sub-agents, integrating the outputs of sub-agents, and ensuring the fulfillment of user needs. Its core functions are as follows: 1) Intelligent question answering based on Agentic RAG (Agent Retrieval Enhanced Generation): For simple user questions, Agentic RAG technology is used to retrieve the knowledge base on demand; the answer is given after integrating the content of the knowledge base; 2) Task decomposition and routing: For complex user task requirements, the task is decomposed into multiple independently executable sub-tasks; based on the requirements of the sub-tasks, the most suitable sub-agent is matched to perform the task according to the capability description of each sub-agent; 3) Sub-agent context management: Based on the sub-task content, the agent generates prompts for calling the sub-agents according to the capability descriptions of the sub-agents and other factors; these prompts may include the outputs of other agents to share and transmit information among multiple agents; in addition, the agent maintains its own historical dialogues with each sub-agent; 4) Sub-agent registration and discovery: The agent maintains a sub-agent registry, recording the capabilities, descriptions, metadata, etc. of each sub-agent for dynamic discovery and invocation. Supports dynamically loading new agents or updating / removing existing agents without service interruption, facilitating system iteration and expansion. 5) Result verification and aggregation: Verifies the execution results of sub-agents, checking for conformity to the expected format and obvious errors, determining whether a retry is necessary; collects the output of each sub-agent, integrates it, and responds to the user with the final task execution result. Parameter configuration agent: User access typically requires configuring access permissions, certificates, IP whitelists, etc. The parameter configuration agent receives the user's description of the configuration in natural language, calls the large model to identify the user's intent, breaks it down into specific tasks such as permission configuration, certificate configuration, and IP whitelist configuration, extracts access terminal ID, access terminal IP, etc., and assembles SQL to input access parameters based on the tasks and information. After parameter configuration is complete, uses the user ID to check if the user access information is fully configured; if missing information is found, it returns missing information and prompts the user to provide the missing content.Sending Agent: Based on the user-provided interface name, message body, and interface specification document, the sending agent constructs a request message for joint debugging and sends it to the joint debugging system. This mainly includes: 1) Converting the request message into a structured JSON message according to the interface specification; 2) Identifying sensitive information in the message body, encrypting it, and digitally signing the message; assembling the message header and message body according to the specification; 3) Calling the general gateway, obtaining the response, and parsing the response code and sensitive information; 4) If the response code is normal, sending the decrypted response back to the collaborative scheduling agent; otherwise, returning the request keyword and reminding the collaborative agent to perform request diagnosis. Diagnostic Agent: The diagnostic agent has pre-built error information and solution cache. It receives the request keyword and request time, queries the joint debugging system's processing logs based on the keyword and time, extracts key information such as the request message and exception logs, compares it with the cached error information, obtains a solution, and sends the solution to the front end. If no solution is found, the response defaults to support information, providing a manual support channel.

[0076] In one embodiment, step S102 specifically includes the following steps:

[0077] The large language model is invoked to perform semantic parsing on the access request, generating an independently executable task decomposition sequence. The task decomposition sequence includes at least one specific configuration task among permission configuration, certificate configuration, and IP whitelist configuration. Based on the configuration task, the access terminal identifier, access terminal address, and access system identification information are extracted from the access request.

[0078] Specifically, the collaborative scheduling agent invokes the parameter agent based on the user's intent and passes the parameters to the parameter agent. The parameter agent assembles SQL to enter the parameter configuration based on the parameters. The parameter agent uses the appId to check whether all the parameters required for joint debugging have been entered. It mainly checks the access terminal interface permissions, access terminal certificate configuration, and access terminal IP whitelist information. For any missing configurations, it returns the parameter information that needs to be configured and requests the access terminal to provide the relevant parameters.

[0079] In one embodiment, after invoking a large language model combined with retrieval enhancement technology to perform semantic parsing on the access request and generating an independently executable task breakdown sequence, the method further includes:

[0080] Based on a pre-built agent registry and vector matching algorithm, the capability requirement features of each subtask in the task decomposition sequence are extracted, and the expert agent nodes that execute the corresponding subtasks are dynamically matched. The addresses of the matched expert agent nodes and the execution parameters corresponding to the subtasks are bound together to generate structured task routing instructions and send them to the corresponding expert agent nodes for execution.

[0081] The collaborative scheduling agent, as the core of the system, provides basic functions such as intelligent question answering, task planning, registration and discovery of sub-agents, and context management, as follows: 1) User intent recognition and task planning: The collaborative agent first calls the LLM (Large Language Model) to classify or decompose the user request. If the request contains only a single task, the Agentic RAG technology is used to read the knowledge base or the corresponding expert agent is called to complete the corresponding task. For example, if the user request is only a knowledge query (such as: how to access a general gateway), the agent calls the Embedding model (Vector Embedding Model) and retrieves relevant content from the vector database, and then calls the LLM to generate an answer. If the request is a complex requirement containing multiple tasks, the collaborative agent will use the LLM to decompose it into a sequence of sub-tasks. For example, if the user request is: "Call the emop.unsign.request interface, the message is...", the collaborative agent recognizes the user intent as sending a transaction and decomposes the transaction sending intent into sub-tasks: Sub-task 1: Construct and send the transaction message; Sub-task 2: If step 1 fails, perform fault diagnosis; The collaborative agent matches and calls the corresponding expert agents in sequence according to the decomposed sub-tasks. 2) Sub-agent Registration and Discovery: The collaborative agent acts as the registry center for all expert agents, maintaining the expert agent registry (JSON file or database table). This table records the capabilities, descriptions, and metadata of each expert agent for dynamic discovery and invocation. The main structure of the registry is as follows: agent_id: unique identifier of the agent; capability_desc: natural language capability description; capability_vec: embedding of the capability description; api_endpoint: invocation address; input_schema: input parameter definition (JSON Schema); output_schema: output parameter definition (optional, JSON Schema). For each sub-task after splitting, the collaborative agent calculates the matching degree between the sub-task and the capability_desc of each expert agent using the cosine similarity of vectors, and selects the expert agent with the highest matching degree to complete the sub-task. 3) Sub-agent Context Management: The collaborative agent uses Redis to store its multi-turn dialogues with the user, as well as its multi-turn dialogues with each sub-agent. When there are too many dialogue rounds (more than 20 rounds), LLM is called to compress the historical dialogue content. When a collaborative agent invokes its sub-agents, it needs to generate specific request parameters based on their input_schema. These request parameters can be structured JSON or prompts in natural language. The specific request parameters are generated by the collaborative agent calling the LLM (Local Management Model). The data may come from historical dialogues between the collaborative agent and the user, or from the output of other agents, to facilitate information sharing and transmission among multiple agents.When the contextual information is insufficient, the collaborative agent proactively queries the user to supplement the data. 4) Dynamic scheduling: After obtaining the output of the sub-agents, the collaborative agent evaluates the sub-agents' outputs and dynamically plans the next task or provides the user with the final answer based on the outputs of each sub-agent. For example, after splitting the task into two sub-tasks: Sub-task 1: Construct and send a transaction message; Sub-task 2: If step 1 fails, perform fault diagnosis; If the fault diagnosis agent handling sub-task 2 returns the result that permissions, certificates, IP whitelists, etc. are not configured, the collaborative agent will ask the user whether relevant configurations need to be added. After obtaining permission, it dynamically adds sub-task 3: parameter configuration and schedules relevant expert agents to complete the task.

[0082] In one embodiment, after the parameters are configured, the following steps are also included:

[0083] The system checks the completeness of parameters; when a parameter is detected to be missing, it returns a parameter supplementation prompt to the user and receives the supplemented parameter information from the user until the parameters are complete.

[0084] In one embodiment, when a missing parameter is detected, a parameter supplementation prompt is returned to the user, and the user's supplemented parameter information is received until the parameter is complete. This specifically includes the following steps:

[0085] Using the access terminal identifier as an index, the system queries the configuration status of permission configuration items, certificate configuration items, and IP whitelist configuration items in the system parameter table to determine whether the access terminal's access information is fully configured. If any of the permission configuration items, certificate configuration items, or IP whitelist configuration items is found to be missing or incomplete, the system determines that the parameters are incomplete and sends a parameter supplementation prompt to the access terminal. The parameter supplementation prompt includes: a list of missing configuration item types, the configuration content requirements and format specifications for each missing item.

[0086] In one embodiment, the method further includes the following steps:

[0087] The system receives parameter information submitted by the user to supplement missing configuration items. The parameter agent performs format verification and validity verification on the supplemented parameters. After the verification is successful, the system updates the parameter table and triggers the parameter completeness check again.

[0088] In one embodiment, complete parameters include: the permission configuration, certificate configuration, and IP whitelist configuration retrieved using the access terminal identifier as an index are all in a configured and verified state.

[0089] In one embodiment, step S102 specifically includes the following steps:

[0090] Sending a joint debugging request message to the access terminal via a sending agent; specifically, sending a joint debugging request message to the access terminal via a sending agent includes: the sending agent converting the request message into a structured message according to the interface specification document; identifying and encrypting sensitive information in the message body, digitally signing the message, assembling the message header and message body according to the interface specification, and generating a joint debugging request message to be sent.

[0091] Specifically, after the parameters are configured, the collaborative scheduling agent selects the sending agent to send a joint debugging request message to the access terminal. Based on the interface name, message body, and interface specification document provided by the user, it constructs the joint debugging request message and sends it to the joint debugging system. This mainly includes: converting the request message into a structured JSON message according to the interface specification; identifying sensitive information in the message body, encrypting it, and digitally signing the message; assembling the message header and message body according to the specification; calling the general gateway, obtaining the general gateway's response, and parsing the response code and sensitive information; if the response code is normal, sending the decrypted response back to the collaborative scheduling agent; otherwise, returning the request keyword and reminding the collaborative agent to perform request diagnosis.

[0092] In one embodiment, receiving feedback information from the access terminal confirming the joint debugging specifically includes: receiving feedback information returned by the access terminal through a general gateway; parsing the response code and sensitive information in the feedback information; when the response code is a normal code, decrypting the sensitive information and sending the decrypted response back to the collaborative scheduling agent; when the response code is an abnormal code, extracting the request keyword and triggering the request diagnosis process.

[0093] In one embodiment, the process of triggering a diagnostic request includes: invoking a diagnostic agent, which receives the request keyword and request time; querying the joint debugging system processing log based on the request keyword and request time, and extracting key information such as the request message and exception log from the processing log; comparing and matching the key information with the error information and solution cache built into the diagnostic agent, obtaining the corresponding solution and feeding it back to the front end for display.

[0094] In one embodiment, the method further includes: when the key information does not match the error information and solution cache, the diagnostic agent returns default information to the front end and provides an entry point for manual support.

[0095] Specifically, the collaborative agent receives the response from the sending agent, determines the request status based on the response code, and if the request fails to process, it calls the diagnostic agent with the request keyword and request time period as parameters. The diagnostic agent queries the request processing log in the joint debugging system based on the request keyword and time period. The diagnostic agent extracts the error information from the log and calls the large model to match the error information and solution. The diagnostic agent has pre-set error messages and solution prompts. The large model finds the corresponding solution based on the prompts and returns it to the diagnostic agent, which then returns the solution to the collaborative agent.

[0096] Figure 3 A schematic diagram of the system access device provided in this application is shown below. Figure 3 As shown in the embodiment of this application, the system access device 300 includes: a receiving module 301, used to receive an access request submitted by an access terminal through a front-end page; a configuration module 302, used to perform intent recognition on the access request, extract key information, and configure the parameters in the key information into a system parameter table through a parameter agent; after the parameter configuration is completed, a joint debugging request message is sent to the access terminal; a sending module 303, used to receive feedback information from the access terminal confirming the joint debugging; wherein, the feedback information includes the content of the joint debugging message; when the feedback information is a confirmation of access, the sending agent sends the content of the joint debugging message to the joint debugging system; and a processing module 304, used to receive feedback information returned by the joint debugging system, determine whether the feedback information is normal; when the feedback information is normal, the system access is confirmed to be successful.

[0097] The system access device 300 provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0098] Figure 4 A schematic diagram of the structure of the electronic device provided in this application. Figure 4 As shown, the electronic device 40 provided in this embodiment includes at least one processor 401 and a memory 402. Optionally, the electronic device 40 further includes a communication component 403. The processor 401, memory 402, and communication component 403 are connected via a bus 404.

[0099] In a specific implementation, at least one processor 401 executes computer execution instructions stored in memory 402, causing at least one processor 401 to perform the above-described method.

[0100] The specific implementation process of processor 401 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0101] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0102] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0103] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0104] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0105] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0106] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0107] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0108] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0109] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0110] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0111] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0112] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0113] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A system access method, characterized in that, include: Receive access requests submitted by access terminals through the front-end page; The access request is subjected to intent recognition, key information is extracted, and the parameters in the key information are configured into the system parameter table through a parameter agent; after the parameter configuration is completed, a joint debugging request message is sent to the access terminal. The system receives feedback information from the access terminal confirming the joint debugging; wherein the feedback information includes the content of the joint debugging message; when the feedback information is confirmation of access, the sending agent sends the content of the joint debugging message to the joint debugging system. The system receives feedback information returned by the joint debugging system and determines whether the feedback information is normal. When the feedback information is normal, the system connection is confirmed to be successful.

2. The method according to claim 1, characterized in that, The process of identifying the intent of the access request and extracting key information specifically includes: The access request is semantically parsed using a large language model to generate an independently executable task breakdown sequence; the task breakdown sequence includes at least one specific configuration task among permission configuration, certificate configuration, and IP whitelist configuration; Based on the configuration task, the access terminal identifier, access terminal address, and access system identifier information are extracted from the access request.

3. The method according to claim 2, characterized in that, After the process of calling a large language model and combining retrieval enhancement technology to perform semantic parsing on the access request and generating an independently executable task decomposition sequence, the process further includes: Based on a pre-set agent registry and vector matching algorithm, the capability requirement features of each subtask in the task decomposition sequence are extracted, and the expert agent node that executes the corresponding subtask is dynamically matched. The matched expert agent node addresses and their corresponding execution parameters are bound together to generate structured task routing instructions, which are then sent to the corresponding expert agent nodes for execution.

4. The method according to claim 2, characterized in that, After the parameters are configured, the following is also included: Check the completeness of the parameters in the system parameter table; When a missing parameter is detected, a parameter supplementation prompt is returned to the user, and the user's supplemented parameter information is received until the parameter is complete.

5. The method according to claim 4, characterized in that, When a missing parameter is detected, a parameter supplementation prompt is returned to the user, and the user's supplemented parameter information is received until the parameter is complete. Specifically, this includes: Using the access terminal identifier as an index, query the configuration status of the permission configuration item, certificate configuration item, and IP whitelist configuration item in the system parameter table to determine whether the access information of the access terminal is fully configured. When any of the permission configuration items, certificate configuration items, or IP whitelist configuration items is detected to be missing or incomplete, the parameters are determined to be incomplete, and a parameter supplementation prompt message is sent to the access terminal. The parameter supplementation prompt message includes: a list of missing configuration item types, configuration content requirements and format specifications corresponding to each missing item.

6. The method according to claim 5, characterized in that, The method further includes: The system receives parameter information submitted by the user to supplement the missing configuration items, and the parameter agent performs format verification and validity verification on the supplemented parameters. After successful verification, the system parameter table is updated and the parameter completeness check is triggered again.

7. The method according to claim 4, characterized in that, The completeness of the parameters includes: the permission configuration, certificate configuration, and IP whitelist configuration retrieved using the access terminal identifier as an index are all configured and verified.

8. The method according to claim 1, characterized in that, The step of initiating a joint debugging request message to the access terminal specifically includes: The intelligent agent sends the joint debugging request message to the access terminal. The step of sending the joint debugging request message to the access terminal via the sending agent specifically includes: According to the interface specification document, the request message is converted into a structured message; sensitive information in the structured message is identified and encrypted, and the structured message is digitally signed. The message header and message body are assembled according to the interface specification to generate a joint debugging request message to be sent.

9. The method according to claim 8, characterized in that, The feedback information received from the access terminal regarding the joint debugging confirmation includes: The system receives feedback information returned by the access terminal through a general gateway. Parse the response code and sensitive information in the feedback information; When the response code is a normal code, the sensitive information is decrypted; When the response code is an error code, the request keyword is extracted and the request diagnosis process is triggered.

10. The method according to claim 9, characterized in that, The trigger request diagnostic process specifically includes: The diagnostic agent is invoked, and the diagnostic agent receives the request keyword and request time. Query the joint debugging system processing log based on the request keyword and the request time, and extract key information such as request message and exception log from the processing log; The key information is compared and matched with the error information and solution cache built into the diagnostic agent to obtain the corresponding solution and display it on the front end.

11. The method according to claim 10, characterized in that, The method further includes: When the key information does not match the error information and solution cache, the diagnostic agent returns default information to the front end.

12. A system access device, characterized in that, include: The receiving module is used to receive access requests submitted by the access terminal through the front-end page; The configuration module is used to identify the intent of the access request, extract key information, and configure the parameters in the key information into the system parameter table through a parameter agent; after the parameter configuration is completed, a joint debugging request message is sent to the access terminal. The sending module is used to receive feedback information from the access terminal confirming the joint debugging; wherein, the feedback information includes the content of the joint debugging message; when the feedback information is confirmation of access, the sending agent sends the content of the joint debugging message to the joint debugging system; The processing module is used to receive feedback information returned by the joint debugging system, determine whether the feedback information is normal, and confirm that the system access is successful when the feedback information is normal.

13. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-11.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-11.

15. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-11.

Images