Intelligent agent interaction method and device between applications, equipment, medium and product

Abstract

The invention discloses an agent interaction method and device between applications, equipment, a medium and a product. The method comprises the following steps: deploying a system service supporting an interoperation protocol between agents on an operating system layer; providing an integrated framework of the system service to a plurality of applications, so that intelligent agents of the applications realize capability access to the system service according to the integrated framework; and after the intelligent agents realize capability access to the system service, intelligent agent interaction between applications is carried out based on forwarding of information between the intelligent agents by the system service. According to the method, the compatibility problem of agent cross-application calling can be solved, and the execution efficiency of complex tasks is improved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, apparatus, device, medium and product for intelligent agent interaction between applications. Background Technology

[0002] With the rapid development of artificial intelligence technology, AI is transitioning from single-point tools to an intelligent agent ecosystem. Intelligent agents mainly consist of four modules: planning and reasoning, memory, tools, and execution. Based on user input, intelligent agents can autonomously break down tasks, call APIs (Application Programming Interfaces), schedule resources across applications, and ultimately achieve the user's expected goals and return results.

[0003] Existing technologies primarily employ the following methods to achieve capability scheduling between intelligent agents in different applications: First, by customizing external services and knowledge graphs within the intelligent agent engine, inter-application intelligent agent scheduling is achieved. However, this approach suffers from high maintenance costs and version dependency issues. Second, by introducing multimodal visual models to simulate clicks on the user application interface, intelligent agent interaction is realized. However, this simulation method is inefficient and suffers from poor stability and difficulty in handling complex logic. Furthermore, neither of these methods effectively solves the problem of dynamic collaboration between intelligent agents; the former suffers from poor compatibility and maintenance difficulties, while the latter lacks reliability. Summary of the Invention

[0004] To address the above technical issues, this application provides a method, apparatus, device, medium, and product for intelligent agent interaction between applications, which can solve the compatibility problem of intelligent agents calling across applications and improve the execution efficiency of complex tasks.

[0005] This application provides a method for intelligent agent interaction between applications, including: Deploy system services that support inter-agent interoperability protocols at the operating system layer; An integration framework for the system services is provided to several applications, enabling the intelligent agents of the applications to access the system services based on the integration framework; After the intelligent agent achieves the ability to access the system service, inter-application intelligent agent interaction is carried out based on the forwarding of information between the intelligent agents by the system service.

[0006] As an improvement to the above solution, the step of enabling inter-application agent interaction based on the forwarding of information between the agents by the system service specifically includes: In response to a demand instruction from a demand-side agent, the system service sends a broadcast based on the demand instruction to at least one service-side agent, so that the service-side agent matching the demand instruction returns a status code. The system service sends the list of service-side intelligent agents corresponding to the status code to the demand-side intelligent agent, so that the demand-side intelligent agent can select the target intelligent agent from the list of service-side intelligent agents to execute the demand instruction; Based on the system service, a communication connection is established between the target agent and the demand agent, so that the demand agent can invoke the target agent to execute the demand instruction.

[0007] As an improvement to the above scheme, in response to a demand instruction from a demand-side agent, the system service sends a broadcast based on the demand instruction to at least one service-side agent, so that service-side agents matching the demand instruction return a status code, including: When a request instruction is received from a requesting agent, the system service sends a broadcast based on the request instruction to at least one service agent, so that the service agent receiving the broadcast can determine whether it has the ability to process the request instruction. If so, it returns a status code to the system service. The system service filters the status codes based on the response conditions in the demand instruction to determine the list of service agent intelligent agents.

[0008] As an improvement to the above scheme, the response conditions include: the longest response time and / or the maximum number of responses; Wherein, if the response condition is the longest response time, the system service starts timing after receiving the demand instruction from the demand agent, and stops timing when the longest response time is reached. The system service determines the list of service agents based on the status codes received within the timing time. If the response condition is the maximum number of responses, the system service counts the received status codes. When the number of status codes is the maximum number of responses, the system service determines the list of service agent intelligent agents based on the received status codes.

[0009] As an improvement to the above solution, the system service sends a list of service-side intelligent agents corresponding to the status code to the demand-side intelligent agent, so that the demand-side intelligent agent can select the target intelligent agent from the list of service-side intelligent agents to execute the demand instruction, including: The system service sends the list of service provider agents corresponding to the status code to the requesting agent, so that the requesting agent can select the target agent to execute the request instruction based on the list of service provider agents using a preset selection algorithm; wherein, the selection algorithm includes, but is not limited to: the fastest response matching algorithm, the optimal capability matching algorithm, or the random selection algorithm.

[0010] As an improvement to the above solution, the step of establishing a communication connection between the target intelligent agent and the demand-side intelligent agent based on the system service, so that the demand-side intelligent agent can invoke the target intelligent agent to execute the demand instruction, includes: A communication connection is established between the target intelligent agent and the demand-side intelligent agent based on the system service; When the system service receives information containing the target agent from the requesting agent, it forwards the information to the target agent so that the target agent can execute the request instruction.

[0011] As an improvement to the above solution, the method further includes: If the demand instruction from the demand-side intelligent agent is a long-running task, then the communication between intelligent agents is carried out through the system service via an asynchronous message interaction framework based on the message push mechanism.

[0012] As an improvement to the above solution, the asynchronous message interaction framework based on the message push mechanism, which enables communication between intelligent agents through the system service, includes: After the demand-side intelligent agent and the service-side intelligent agent establish a communication connection based on the system service, the message push interface of the demand-side intelligent agent is provided to the service-side intelligent agent, so that after the service-side intelligent agent completes the demand instruction of the demand-side intelligent agent, it calls the message push interface to push the result to the demand-side intelligent agent.

[0013] As an improvement to the above solution, the integration framework that provides the system services is provided to several applications, enabling the intelligent agents of the applications to access the system services based on the integration framework, including: The system service integration framework is output to several applications in the form of a software development kit (SDK) so that the SDK can be integrated into the intelligent agents of several applications, enabling the intelligent agents of the applications to push and pull messages of the system service based on the SDK.

[0014] As an improvement to the above solution, the communication between the system service and the intelligent agent is implemented based on the Android system interface definition language AIDL.

[0015] As an improvement to the above scheme, before the inter-application agent interaction is performed based on the forwarding of information between the agents according to the system service, the method further includes: In response to the binding service request of the intelligent agent, after the system service binds with the intelligent agent, it returns the invoked proxy object to the intelligent agent; wherein, the proxy object is used to connect to the service; In response to the agent's registration callback interface operation, the system service adds the callback interface to a thread-safe container to register a listener; wherein, the callback interface is used to allow the system service to send information to the agent; The system service sets up a death proxy for the callback interface to monitor the process of the agent. When the process of the agent terminates, the binding between the agent and the system service is released.

[0016] This application also provides an inter-application intelligent agent interaction device, including: The system service module is used to deploy system services that support inter-agent interoperability protocols at the operating system layer; A capability integration module is used to provide an integration framework for the system services to several applications, so that the intelligent agents of the applications can access the capabilities of the system services according to the integration framework. The information interaction module is used to perform inter-application agent interaction based on the forwarding of information between agents by the system service after the agent has achieved access to the system service.

[0017] Furthermore, the process of enabling inter-application agent interaction based on the forwarding of information between the agents by the system service specifically includes: In response to a demand instruction from a demand-side agent, the system service sends a broadcast based on the demand instruction to at least one service-side agent, so that the service-side agent matching the demand instruction returns a status code. The system service sends the list of service-side intelligent agents corresponding to the status code to the demand-side intelligent agent, so that the demand-side intelligent agent can select the target intelligent agent from the list of service-side intelligent agents to execute the demand instruction; Based on the system service, a communication connection is established between the target agent and the demand agent, so that the demand agent can invoke the target agent to execute the demand instruction.

[0018] Further, in response to a demand instruction from a demand-side agent, the system service sends a broadcast based on the demand instruction to at least one service-side agent, causing the service-side agent matching the demand instruction to return a status code, including: When a request instruction is received from a requesting agent, the system service sends a broadcast based on the request instruction to at least one service agent, so that the service agent receiving the broadcast can determine whether it has the ability to process the request instruction. If so, it returns a status code to the system service. The system service filters the status codes based on the response conditions in the demand instruction to determine the list of service agent intelligent agents.

[0019] Furthermore, the response conditions include: the longest response time and / or the maximum number of responses; Wherein, if the response condition is the longest response time, the system service starts timing after receiving the demand instruction from the demand agent, and stops timing when the longest response time is reached. The system service determines the list of service agents based on the status codes received within the timing time. If the response condition is the maximum number of responses, the system service counts the received status codes. When the number of status codes is the maximum number of responses, the system service determines the list of service agent intelligent agents based on the received status codes.

[0020] Further, the system service sends the list of service-side intelligent agents corresponding to the status code to the demand-side intelligent agent, so that the demand-side intelligent agent can select the target intelligent agent from the list of service-side intelligent agents to execute the demand instruction, including: The system service sends the list of service provider agents corresponding to the status code to the requesting agent, so that the requesting agent can select the target agent to execute the request instruction based on the list of service provider agents using a preset selection algorithm; wherein, the selection algorithm includes, but is not limited to: the fastest response matching algorithm, the optimal capability matching algorithm, or the random selection algorithm.

[0021] Furthermore, establishing a communication connection between the target agent and the demand agent based on the system service, so that the demand agent can invoke the target agent to execute the demand instruction, includes: A communication connection is established between the target intelligent agent and the demand-side intelligent agent based on the system service; When the system service receives information containing the target agent from the requesting agent, it forwards the information to the target agent so that the target agent can execute the request instruction.

[0022] Furthermore, the device is also used for: If the demand instruction from the demand-side intelligent agent is a long-running task, then the communication between intelligent agents is carried out through the system service via an asynchronous message interaction framework based on the message push mechanism.

[0023] Furthermore, the asynchronous message interaction framework based on the message push mechanism, which enables communication between intelligent agents through the system service, includes: After the demand-side intelligent agent and the service-side intelligent agent establish a communication connection based on the system service, the message push interface of the demand-side intelligent agent is provided to the service-side intelligent agent, so that after the service-side intelligent agent completes the demand instruction of the demand-side intelligent agent, it calls the message push interface to push the result to the demand-side intelligent agent.

[0024] Furthermore, the integration framework providing the system services is provided to several applications, enabling the intelligent agents of the applications to access the system services based on the integration framework, including: The system service integration framework is output to several applications in the form of a software development kit (SDK) so that the SDK can be integrated into the intelligent agents of several applications, enabling the intelligent agents of the applications to push and pull messages of the system service based on the SDK.

[0025] Furthermore, the communication between the system service and the intelligent agent is implemented based on the Android system interface definition language AIDL.

[0026] Furthermore, the device is also used for: In response to the binding service request of the intelligent agent, after the system service binds with the intelligent agent, it returns the invoked proxy object to the intelligent agent; wherein, the proxy object is used to connect to the service; In response to the agent's registration callback interface operation, the system service adds the callback interface to a thread-safe container to register a listener; wherein, the callback interface is used to allow the system service to send information to the agent; The system service sets up a death proxy for the callback interface to monitor the process of the agent. When the process of the agent terminates, the binding between the agent and the system service is released.

[0027] This application also provides a computer device, including a processor and a memory, wherein the memory stores a computer program and the computer program is configured to be executed by the processor, wherein the processor executes the computer program to implement the inter-application intelligent agent interaction method described in any of the preceding claims.

[0028] This application also provides a computer-readable storage medium storing a computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the inter-application agent interaction method described above.

[0029] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the inter-application intelligent agent interaction method described in any of the above claims.

[0030] Compared to existing technologies, the beneficial effects of the inter-application intelligent agent interaction method, apparatus, device, medium, and product provided in this application are as follows: By deploying system services that support inter-application interoperability protocols at the operating system layer, and providing an integration framework for the system services to several applications, the intelligent agents of the applications can access the system services based on the integration framework. After the intelligent agents access the system services, inter-application intelligent agent interaction is performed based on the forwarding of information between the intelligent agents by the system services. This solves the compatibility and maintenance problems of cross-application calls in the prior art, improves the determinism and efficiency of complex task execution between application intelligent agents, and enhances the user experience. Attached Figure Description

[0031] Figure 1 This is a flowchart illustrating an inter-application intelligent agent interaction method provided in an embodiment of this application; Figure 2 This is a timing flowchart of inter-application intelligent agent interaction provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an inter-application intelligent agent interaction device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0032] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0033] Please see Figure 1 , Figure 1 This is a flowchart illustrating an inter-application intelligent agent interaction method provided in an embodiment of this application. The inter-application intelligent agent interaction method includes: S1: Deploy system services that support inter-agent interoperability protocols at the operating system layer; S2: Provide an integration framework for the system services to several applications, so that the intelligent agents of the applications can access the system services based on the integration framework; S3: After the intelligent agent achieves the ability to access the system service, based on the forwarding of information between the intelligent agents by the system service, intelligent agent interaction between applications is carried out.

[0034] Specifically, the embodiments of this application are applied to cloud phone systems, but can also be applied to mobile phone systems of terminal devices. For the interaction between intelligent agents of various applications in a mobile phone system, a system service supporting inter-agent interoperability protocols is deployed within the cloud phone's operating system to achieve standardized communication between application intelligent agents. Specifically, a system service that does not require direct user interaction is developed in the operating system background and set to run resident in the system.

[0035] Interoperability protocols between intelligent agents can be A2A protocols. A2A (Agent to Agent) protocols are open protocols launched by Google for communication between intelligent agents. They can promote interoperability between different AI (Artificial Intelligence) intelligent agents and allow intelligent agents to communicate and collaborate effectively in dynamic, multi-agent ecosystems.

[0036] This application's embodiments create a central engine service (i.e., a system service) based on the A2A protocol. Using the standard JSON (JavaScript Object Notation) communication protocol, it is output to applications via an SDK (Software Development Kit). Application clients using the SDK implement message push and retrieval from the system service, thereby facilitating communication between intelligent agents. Furthermore, after each application interacts with the system service, service binding is established, integrating the ability to call the system service into the intelligent agents of several applications. This allows intelligent agents to invoke the capabilities of other intelligent agents through the system service to complete target tasks. In other words, intelligent agent interaction between applications is based on the system service's forwarding of information between intelligent agents.

[0037] This application's embodiments provide a standardized interoperability foundation for intelligent agents in different applications by deploying a unified system service at the operating system layer, enabling cross-application intelligent agent interconnection. By establishing a system service, intelligent agents in each application only need to integrate the ability to call this system service, without needing to directly adapt to each other, thus reducing the complexity of interactions between intelligent agents.

[0038] As one optional embodiment, the integration framework providing the system services is provided to several applications, enabling the intelligent agents of the applications to access the system services based on the integration framework, including: The system service integration framework is output to several applications in the form of a software development kit (SDK) so that the SDK can be integrated into the intelligent agents of several applications, enabling the intelligent agents of the applications to push and pull messages of the system service based on the SDK.

[0039] Specifically, to enable third-party applications to easily access the intelligent agent interaction ecosystem of this application, the system service provides a unified Client SDK (Client Software Development Kit), i.e., an integration framework. The Client SDK is a pre-packaged code toolkit used for integration into client applications, allowing applications to easily and quickly use specific functions or services provided by the system service. In this embodiment, after deploying the system service, specifically within the cloud phone's capability open platform, a corresponding SDK is provided for third-party applications to integrate their intelligent agents. Each client application uses the system service's Client SDK to integrate its ability to call the system service into the intelligent agent service.

[0040] The SDK includes: all interface files defined for communicating with system services; utility classes encapsulating complex processes such as binding, registering, sending commands, receiving broadcasts, and callbacks with system services; API documentation and integration guidelines. After obtaining the SDK, the application agent integrates by initializing the client utility classes provided by the SDK in the agent's relevant code, thus completing the connection between the agent and the underlying system services.

[0041] This application provides a standardized software development kit, enabling applications to quickly acquire the ability to interact across applications with their intelligent agents at a relatively low integration cost. This greatly reduces the development threshold and integration cost, which is conducive to the promotion of the technology and the construction of the ecosystem.

[0042] As one optional embodiment, the communication between the system service and the intelligent agent is implemented based on the Android system interface definition language AIDL.

[0043] Specifically, this application embodiment deploys system services based on the A2A protocol in a mobile operating system based on AIDL (Android System Interface Definition Language). AIDL is an interface definition language used by the Android system to implement IPC (Inter-Process Communication). It defines the interaction rules between different processes by declaring interface methods, solving the problem that processes cannot directly share memory. By defining the communication interface language through AIDL, two applications can call functions and transfer data to each other.

[0044] The system service in this application inherits the AIDL service and enhances the intelligent agent communication service capabilities, including extracting and storing information from broadcasts that conform to the format protocol when receiving broadcasts, sending broadcast notification information to application intelligent agents that have integrated services in the operating system, and completing end-to-end matching and relay transmission.

[0045] The process of information interaction between application agents through AIDL services includes: the demand-side agent and the service-side agent respectively bind to the AIDL-based system service; the system service registers a listener for the service-side agent to monitor its process; the demand-side agent sends a broadcast to the system service, which then distributes the broadcast to the service-side agent; after the connection is established, the demand-side agent and the service-side agent communicate through the system service.

[0046] The embodiments of this application implement communication based on the mature AIDL mechanism of the Android system, making full use of the stable and efficient cross-process communication capabilities provided by the underlying operating system, and ensuring the reliability and performance of communication between system services and various application intelligent agents.

[0047] As one optional embodiment, before the inter-application agent interaction is performed based on the forwarding of information between the agents according to the system service, the method further includes: In response to the binding service request of the intelligent agent, after the system service binds with the intelligent agent, it returns the invoked proxy object to the intelligent agent; wherein, the proxy object is used to connect to the service; In response to the agent's registration callback interface operation, the system service adds the callback interface to a thread-safe container to register a listener; wherein, the callback interface is used to allow the system service to send information to the agent; The system service sets up a death proxy for the callback interface to monitor the process of the agent. When the process of the agent terminates, the binding between the agent and the system service is released.

[0048] Specifically, in the process of information interaction between application agents through AIDL-based system services, each agent first binds to the service and registers a listener. Each agent sends a service binding request to the system service. In response to the agent's binding request, the system service establishes a cross-process connection with the agent through an inter-process communication mechanism. Specifically, the agent creates an explicit Intent (specifying the service package name and class name) and calls the bindService() method to bind the service. The system service establishes the cross-process connection through the Binder (binder) driver. Binder is an efficient and secure inter-process communication mechanism provided by the Android system. As a low-level tool, it allows different processes to access services and data as if they were local calls.

[0049] Furthermore, once the system service and the agent successfully connect, the agent obtains a proxy object from the system service that can directly call methods to connect to the service. Specifically, after a successful connection, the system service returns an IBinder interface object to the agent, which then obtains an IBroadcastService (broadcast service interface) proxy object to establish a connection to the service.

[0050] Furthermore, the service-side agent registers a callback interface, and the system service adds the callback interface to a thread-safe container to register a listener. The callback interface is used by the system service to send information to the service-side agent. Specifically, the service-side agent creates a stub implementation instance of the callback interface, calls the `registerListener()` method to register the callback interface, and the system service adds the callback interface to a thread-safe container, thus registering the listener. It should be noted that the demand-side agent does not register a listener; it only obtains a proxy object for sending broadcasts.

[0051] Furthermore, the system service sets up a death proxy for the callback interface to monitor the process of the service-side intelligent agent. When the process of the service-side intelligent agent terminates, the binding between the service-side intelligent agent and the system service is released. The system service listens to whether the intelligent agent's process is alive, and automatically cleans up resources when the intelligent agent's process terminates to avoid anomalies.

[0052] This application's embodiments, through service binding, callback registration, and setting death proxies, enable the system service to establish a stable connection with each intelligent agent and monitor its process status in real time. When an intelligent agent's process terminates abnormally, the system can automatically detect and clean up related resources, improving the stability and maintainability of the entire interactive system.

[0053] As one optional embodiment, the inter-application agent interaction based on the forwarding of information between the agents by the system service specifically includes: In response to a demand instruction from a demand-side agent, the system service sends a broadcast based on the demand instruction to at least one service-side agent, so that the service-side agent matching the demand instruction returns a status code. The system service sends the list of service-side intelligent agents corresponding to the status code to the demand-side intelligent agent, so that the demand-side intelligent agent can select the target intelligent agent from the list of service-side intelligent agents to execute the demand instruction; Based on the system service, a communication connection is established between the target agent and the demand agent, so that the demand agent can invoke the target agent to execute the demand instruction.

[0054] Specifically, the request-side agent sends a broadcast to the system service, which then distributes the broadcast to the service-side agents. First, the request-side agent creates a data packet containing the action type and data payload, and serializes it. Then, it transmits the call request to the system service process via the Binder driver. The request-side agent packages the serialized data and sends it to the system service. The system service checks the request-side agent's calling permissions and verifies the digital signature to prevent forged requests. Upon receiving the data, the system service deserializes it to obtain a data packet containing the action type and data payload, enabling non-blocking asynchronous calls so that the system service can distribute the message to each service-side agent. During this process, CopyOnWriteArrayList is used to ensure thread safety during iteration.

[0055] When a demand-side intelligent agent performs capability scheduling, it generates a unique event identifier and sends a demand command to the system service. In response to the demand command, the system service returns response data to the demand-side intelligent agent and then forwards a broadcast based on the demand command to several service-side intelligent agents with integrated SDKs to find target intelligent agents that meet the demand command. Upon receiving the broadcast, each service-side intelligent agent determines whether it has the capability to complete the demand command. If a service-side intelligent agent matches the demand command, it returns a status code to the system service. Based on the received status codes, the system service compiles a list of service-side intelligent agents that meet the demand and sends this list to the demand-side intelligent agent. The demand-side intelligent agent then selects one or more service-side intelligent agents as target intelligent agents to execute the demand command, and establishes a communication connection between the demand-side intelligent agent and the target intelligent agent through the system service. Through dialogue and response between the demand-side intelligent agent and the target intelligent agent, the corresponding demand command is completed. After the service-side intelligent agent completes the corresponding operation, the communication round ends.

[0056] For example, please refer to Figure 2 , Figure 2This is a timing flowchart of inter-application intelligent agent interaction provided in an embodiment of this application. After the requesting intelligent agent, the service intelligent agent, and the system service are bound together, the requesting intelligent agent sends a request command (i.e., send Command) to the system service. The request command includes a unique event identifier (eventId), a maximum response time (timeout), and a maximum number of responses (max). Here, timeout = 5000 ms indicates a maximum response time of 5 seconds; max = 2 indicates a maximum number of responses from the service intelligent agent (2). After receiving the request command, the system service returns response data (ASK) to the requesting intelligent agent, indicating that the system service has received the request. Then, the system service broadcasts the relevant information of the request command (intent(eventId)) to each service intelligent agent. Each service intelligent agent assesses its own capabilities. If a service intelligent agent has the ability to execute the request command, it returns a status code "ready" to the system service. The status code includes the unique event identifier (eventId) and the identifier of the service intelligent agent (client_b). Based on the received status code, the system service sends the corresponding list of service intelligent agents (candidates[client_b]) to the requesting intelligent agent. The requesting agent selects a target agent from the list of service agents (i.e., choose(client_b)) and notifies the system service, so that the system service forwards the requesting agent's payload information to the target agent (i.e., forward(A's payload)), which is service agent 1 in the diagram. Then, a communication connection is established between the two agents based on the system service. Finally, when the task of the request instruction is completed, the service agent returns the completion information finish(eventId) to the system service to end the communication.

[0057] This application embodiment broadcasts messages based on system services, which can dynamically discover available service agents in the current environment and autonomously select the most suitable agent for task delegation based on the returned list. This process is coordinated by the system service, which not only realizes the automatic matching and optimization of service resources, but also ensures the reliability of the communication link for subsequent calls, significantly improving the efficiency and success rate of cross-agent task execution.

[0058] As one optional embodiment, in response to a demand instruction from a demand-side agent, the system service sends a broadcast based on the demand instruction to at least one service-side agent, causing the service-side agent matching the demand instruction to return a status code, including: When a request instruction is received from a requesting agent, the system service sends a broadcast based on the request instruction to at least one service agent, so that the service agent receiving the broadcast can determine whether it has the ability to process the request instruction. If so, it returns a status code to the system service. The system service filters the status codes based on the response conditions in the demand instruction to determine the list of service agent intelligent agents.

[0059] Specifically, after the requesting agent generates a unique event identifier and sends a request command to the system service, in response to the requesting agent's request command, the system service extracts the information from the request command, performs parameter verification, records the corresponding unique event identifier and agent information, and then returns response data to the requesting agent, indicating that the system service has received the request and starts a timeout countdown. Simultaneously, a broadcast based on the request command is sent to several service-side agents that have integrated the SDK. The request command specifies the response conditions used to filter agents, such as limiting the number of agents that respond within a preset time; for example, if two agents respond within 5 seconds, agent matching is interrupted.

[0060] For example, when a requesting agent needs to schedule the capabilities of other agents, it sends a request instruction to the system service. This request instruction includes: a unique event identifier for this session, used for receiving broadcasts later; a list of candidate service agents; the name of the target agent, which is left blank during the initial communication and filled in after a successful agent match, enabling direct proxying and rapid flow of subsequent communication on the same event; a maximum response time to limit the response duration; a maximum number of responses to limit the number of service agents that can respond; and the main content of the request instruction, which the agent analyzes to initiate a dialogue.

[0061] The status code of the service agent includes: a unique event identifier for this session; the identifier of the service agent; the status of the service agent; and a declaration of the service agent's optional capabilities. When the system service receives the status code returned by the service agent, it filters the service agents that returned status codes based on response conditions, such as the maximum number of responses an agent can provide, to obtain a list of service agents, which is then notified to the requesting agent.

[0062] This application embodiment provides a basis for traceability and timeliness management of the entire interaction process by verifying and recording the demand instructions and sending a receipt to start the timer. By requiring the service agent to judge its own capabilities and return a status code, the system service filters the requests, which clarifies the responsibilities of each party in the service discovery process, makes the filtering logic clearer, and enhances the standardization and controllability of the interaction process.

[0063] As one optional embodiment, the response conditions include: the longest response time and / or the maximum number of responses; Wherein, if the response condition is the longest response time, the system service starts timing after receiving the demand instruction from the demand agent, and stops timing when the longest response time is reached. The system service determines the list of service agents based on the status codes received within the timing time. If the response condition is the maximum number of responses, the system service counts the received status codes. When the number of status codes is the maximum number of responses, the system service determines the list of service agent intelligent agents based on the received status codes.

[0064] Specifically, the system service can limit the responses of other service agents based on the longest response time and / or the maximum number of responses. When the longest response time is used as the response condition, after receiving the request instruction from the requesting agent, the system service verifies and records the request instruction and returns response data to the requesting agent, so that the requesting agent and the system service start timing simultaneously. When the timing reaches the longest response time, the timing stops. At the start of the timing, the system service sends a broadcast based on the request instruction to at least one service agent, so that the receiving service agent can determine whether it has the ability to process the request instruction. If so, it returns a status code to the system service. At the end of the timing, the system service compiles the service agents corresponding to the received status codes into a service agent list and sends it to the requesting agent.

[0065] When the maximum number of responses is used as the response condition, after the system service sends a broadcast based on the demand command to at least one service agent, it starts recording the number of status codes received. When the number of received status codes reaches the maximum number of responses, it stops receiving status codes and organizes the service agents corresponding to the currently received status codes into a list of service agents, which is then sent to the demand agent.

[0066] When the response conditions include the longest response time and the maximum number of responses, after receiving the request instruction from the requesting agent, the system service verifies and records the instruction and returns response data to the requesting agent, so that the requesting agent and the system service start timing simultaneously. During the timing process, if the number of received status codes reaches the maximum number of responses, the system service stops receiving status codes and compiles the service agents corresponding to the received status codes into a service agent list. If the system service receives status codes but does not reach the maximum number of responses when the timing reaches the longest response time, the system service stops timing and status code reception and compiles the service agents corresponding to the received status codes into a service agent list.

[0067] For example, depending on actual needs, the fastest response can be obtained by limiting the maximum number of responses to 1, or by setting a maximum response time to select from multiple responding service agents. Furthermore, if no matching agent responds within the response conditions, the system service directly replies to the requesting agent that no corresponding service agent can fulfill the request instruction, thus prompting the user to manually or through an agent to complete the request in other ways.

[0068] This application embodiment balances efficiency and resource consumption by setting two key parameters: the longest response time and the maximum number of responses. The longest response time avoids indefinite waiting and ensures the overall response speed of the system; the maximum number of responses controls the scale of competing service providers, thereby dynamically adjusting the response speed according to the actual load and improving efficiency.

[0069] As one optional embodiment, the system service sends a list of service-side intelligent agents corresponding to the status code to the demand-side intelligent agent, so that the demand-side intelligent agent selects the target intelligent agent from the list of service-side intelligent agents to execute the demand instruction, including: The system service sends the list of service provider agents corresponding to the status code to the requesting agent, so that the requesting agent can select the target agent to execute the request instruction based on the list of service provider agents using a preset selection algorithm; wherein, the selection algorithm includes, but is not limited to: the fastest response matching algorithm, the optimal capability matching algorithm, or the random selection algorithm.

[0070] Specifically, after the system service sends the list of service provider agents to the demand agent, the demand agent can select from the candidate service provider agents using the fastest response matching algorithm, the optimal capability matching algorithm, or the random selection algorithm. Based on the algorithm, the target agent for executing the demand command is determined, thereby establishing a communication connection between the demand agent and the target agent.

[0071] The fastest response matching algorithm selects the target agent based on the time it takes for the service agent to return a status code. The optimal capability matching algorithm considers the available capabilities of the service agent in its returned status code, evaluating its capabilities and prioritizing the service agent with the highest capability as the target agent. The random selection algorithm randomly selects a predetermined number of service agents from the list as target agents.

[0072] This application provides a variety of preset selection algorithms, allowing the demand-side intelligent agent to select the most suitable target intelligent agent for the current scenario based on the real-time requirements, quality requirements, or load balancing considerations of the specific task. This improves the flexibility of intelligent agent selection, optimizes the overall task execution efficiency, and achieves more refined scheduling.

[0073] As one optional embodiment, establishing a communication connection between the target agent and the demand agent based on the system service, so that the demand agent can invoke the target agent to execute the demand instruction, includes: A communication connection is established between the target intelligent agent and the demand-side intelligent agent based on the system service; When the system service receives information containing the target agent from the requesting agent, it forwards the information to the target agent so that the target agent can execute the request instruction.

[0074] Specifically, after identifying the target agent, a communication connection is established between the requesting agent and the target agent through a system service. Multiple conversations between the requesting agent and the target agent are achieved based on the system service's forwarding of information between the agents. In subsequent broadcasts, the requesting agent can achieve point-to-point proxy forwarding as long as the target agent name in the request command contains information about the service agent. The requesting agent and the target agent engage in dialogue and response until the requesting agent completes its request command. At this point, the requesting agent sends a completion response message to the system service, causing the system service to terminate the current communication.

[0075] For example, the demand-side intelligent agent and the target intelligent agent communicate point-to-point through the main content of the demand instruction. For instance, the demand-side intelligent agent sends the instruction "I want to order a spicy chicken delivery." Upon receiving this instruction, the target intelligent agent generates the order and then responds with card information and a Scheme (protocol header) to the demand-side intelligent agent. The demand-side intelligent agent interacts with the user by displaying the card information. When the user clicks to confirm, the Scheme initiates the order from the target intelligent agent, allowing the user to pay for the order, thus completing the closed loop of the entire demand chain and enabling intelligent interaction between intelligent agents across applications.

[0076] The embodiments of this application establish communication connections through system services, which simplifies the complexity of direct communication between intelligent agents, while ensuring the reliability and accuracy of instruction transmission.

[0077] As one optional embodiment, the method further includes: If the demand instruction from the demand-side intelligent agent is a long-running task, then the communication between intelligent agents is carried out through the system service via an asynchronous message interaction framework based on the message push mechanism.

[0078] Specifically, for long-running tasks, an asynchronous message interaction framework is used between intelligent agents to collaborate, leveraging the message push mechanism. After the requesting intelligent agent initiates a request, the service-side intelligent agent immediately confirms and generates a task credential; once the task is completed, the service-side intelligent agent proactively pushes the result to the requesting intelligent agent through an agreed-upon callback interface, thus achieving asynchronous message push.

[0079] This application's embodiments introduce a message-push-based asynchronous interaction framework for long-running tasks. This allows the requesting agent to continue processing other tasks without being blocked after initiating a task, thus improving the system's concurrent processing capabilities and overall responsiveness. This interaction method is not only capable of handling immediate tasks but also efficiently supports background jobs that require long-running or waiting periods.

[0080] As one optional embodiment, the asynchronous message interaction framework based on the message push mechanism, which enables communication between intelligent agents through the system service, includes: After the demand-side intelligent agent and the service-side intelligent agent establish a communication connection based on the system service, the message push interface of the demand-side intelligent agent is provided to the service-side intelligent agent, so that after the service-side intelligent agent completes the demand instruction of the demand-side intelligent agent, it calls the message push interface to push the result to the demand-side intelligent agent.

[0081] Specifically, after the demand-side intelligent agent and the target intelligent agent establish a communication connection and confirm the process, the demand-side intelligent agent submits a request to the service-side intelligent agent through the system service, and pre-sets a message push API for receiving results. Upon receiving the request, the service-side intelligent agent returns a task receipt confirmation message. Then, the service-side intelligent agent executes the task in the background. Once completed, it calls the message push API provided by the demand-side intelligent agent according to the pre-set callback interface, sending the task processing result to the demand-side intelligent agent, thus achieving asynchronous push. After receiving the result, the demand-side intelligent agent's push service updates its internal state and notifies the user to receive the corresponding result or trigger subsequent processes.

[0082] This application embodiment provides the message push interface of the demand-side intelligent agent to the service-side intelligent agent. After the task is completed, the service-side intelligent agent actively pushes the result. The demand-side intelligent agent does not need to perform inefficient polling. Through this asynchronous push mechanism, unnecessary network communication and resource consumption are reduced, and the real-time performance of asynchronous interaction and overall system efficiency are improved.

[0083] This application deploys a system service that supports inter-agent interoperability protocols at the operating system layer, and provides an integration framework for the system service to several applications. This enables the agents of the applications to access the system service based on the integration framework. After the agents access the system service, inter-application agent interaction is performed based on the forwarding of information between the agents by the system service. This solves the compatibility and maintenance problems of cross-application calls in the prior art, improves the determinism and efficiency of complex task execution between application agents, and enhances the user experience.

[0084] Accordingly, this application also provides an inter-application intelligent agent interaction device, which can implement all the processes of the inter-application intelligent agent interaction method in the above embodiments.

[0085] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of an inter-application intelligent agent interaction device provided in an embodiment of this application. The inter-application intelligent agent interaction device includes: System service module 301 is used to deploy system services that support inter-agent interoperability protocols at the operating system layer; The capability integration module 302 is used to provide an integration framework for the system services to several applications, so that the intelligent agents of the applications can access the capabilities of the system services according to the integration framework. The information interaction module 303 is used to perform inter-application agent interaction based on the forwarding of information between agents by the system service after the agent has achieved the ability to access the system service.

[0086] Preferably, the step of performing inter-application agent interaction based on the forwarding of information between the agents by the system service specifically includes: In response to a demand instruction from a demand-side agent, the system service sends a broadcast based on the demand instruction to at least one service-side agent, so that the service-side agent matching the demand instruction returns a status code. The system service sends the list of service-side intelligent agents corresponding to the status code to the demand-side intelligent agent, so that the demand-side intelligent agent can select the target intelligent agent from the list of service-side intelligent agents to execute the demand instruction; Based on the system service, a communication connection is established between the target agent and the demand agent, so that the demand agent can invoke the target agent to execute the demand instruction.

[0087] Preferably, in response to a demand instruction from a demand-side agent, the system service sends a broadcast based on the demand instruction to at least one service-side agent, causing service-side agents matching the demand instruction to return a status code, including: When a request instruction is received from a requesting agent, the system service sends a broadcast based on the request instruction to at least one service agent, so that the service agent receiving the broadcast can determine whether it has the ability to process the request instruction. If so, it returns a status code to the system service. The system service filters the status codes based on the response conditions in the demand instruction to determine the list of service agent intelligent agents.

[0088] Preferably, the response conditions include: the longest response time and / or the maximum number of responses; Wherein, if the response condition is the longest response time, the system service starts timing after receiving the demand instruction from the demand agent, and stops timing when the longest response time is reached. The system service determines the list of service agents based on the status codes received within the timing time. If the response condition is the maximum number of responses, the system service counts the received status codes. When the number of status codes is the maximum number of responses, the system service determines the list of service agent intelligent agents based on the received status codes.

[0089] Preferably, the system service sends a list of service-side intelligent agents corresponding to the status code to the demand-side intelligent agent, so that the demand-side intelligent agent selects the target intelligent agent from the list of service-side intelligent agents to execute the demand instruction, including: The system service sends the list of service provider agents corresponding to the status code to the requesting agent, so that the requesting agent can select the target agent to execute the request instruction based on the list of service provider agents using a preset selection algorithm; wherein, the selection algorithm includes, but is not limited to: the fastest response matching algorithm, the optimal capability matching algorithm, or the random selection algorithm.

[0090] Preferably, establishing a communication connection between the target agent and the demand agent based on the system service, so that the demand agent can invoke the target agent to execute the demand instruction, includes: A communication connection is established between the target intelligent agent and the demand-side intelligent agent based on the system service; When the system service receives information containing the target agent from the requesting agent, it forwards the information to the target agent so that the target agent can execute the request instruction.

[0091] Preferably, the device is further used for: If the demand instruction from the demand-side intelligent agent is a long-running task, then the communication between intelligent agents is carried out through the system service via an asynchronous message interaction framework based on the message push mechanism.

[0092] Preferably, the asynchronous message interaction framework based on the message push mechanism enables communication between intelligent agents through the system service, including: After the demand-side intelligent agent and the service-side intelligent agent establish a communication connection based on the system service, the message push interface of the demand-side intelligent agent is provided to the service-side intelligent agent, so that after the service-side intelligent agent completes the demand instruction of the demand-side intelligent agent, it calls the message push interface to push the result to the demand-side intelligent agent.

[0093] Preferably, the integration framework providing the system services is provided to several applications, enabling the intelligent agents of the applications to access the system services based on the integration framework, including: The system service integration framework is output to several applications in the form of a software development kit (SDK) so that the SDK can be integrated into the intelligent agents of several applications, enabling the intelligent agents of the applications to push and pull messages of the system service based on the SDK.

[0094] Preferably, the communication between the system service and the intelligent agent is implemented based on the Android system interface definition language AIDL.

[0095] Preferably, the device is further used for: In response to the binding service request of the intelligent agent, after the system service binds with the intelligent agent, it returns the invoked proxy object to the intelligent agent; wherein, the proxy object is used to connect to the service; In response to the agent's registration callback interface operation, the system service adds the callback interface to a thread-safe container to register a listener; wherein, the callback interface is used to allow the system service to send information to the agent; The system service sets up a death proxy for the callback interface to monitor the process of the agent. When the process of the agent terminates, the binding between the agent and the system service is released.

[0096] In specific implementation, the working principle, control process and technical effects of the inter-application intelligent agent interaction device provided in this application embodiment are the same as the inter-application intelligent agent interaction method in the above embodiment, and will not be repeated here.

[0097] See Figure 4 , Figure 4This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. The computer device includes: a processor 401, a memory 402, and a computer program stored in the memory 402 and executable on the processor 401. When the processor 401 executes the computer program, it implements the steps in the above-described embodiments of the intelligent agent interaction method between applications. Alternatively, when the processor 401 executes the computer program, it implements the functions of each module / unit in the above-described device embodiments.

[0098] For example, the computer program may be divided into one or more modules / units, which are stored in the memory 402 and executed by the processor 401 to complete this application. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the computer device.

[0099] The computer device may include, but is not limited to, processor 401 and memory 402. Those skilled in the art will understand that the schematic diagram is merely an example of a computer device and does not constitute a limitation on the computer device. It may include more or fewer components than illustrated, or combine certain components, or different components. For example, the computer device may also include input / output devices, network access devices, buses, etc.

[0100] The processor 401 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor 401 is the control center of the computer device, connecting various parts of the entire computer device through various interfaces and lines.

[0101] The memory 402 can be used to store the computer programs and / or modules. The processor 401 implements various functions of the computer device by running or executing the computer programs and / or modules stored in the memory 402 and calling the data stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function, etc.; the data storage area may store data created based on the use of the mobile phone, etc. In addition, the memory 402 may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0102] Wherein, if the modules / units integrated into the computer device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by the processor 401, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms.

[0103] This application also provides a computer-readable storage medium, which includes a stored computer program, wherein the computer program, when running, controls the device where the computer-readable storage medium is located to execute the inter-application agent interaction method described in any of the above embodiments.

[0104] This application also provides a computer program product, including a computer program / instruction, which, when executed by the computer program / instruction processor, implements the inter-application intelligent agent interaction method described in any of the above embodiments.

[0105] The methods in this application can be implemented, in whole or in part, by software, hardware, firmware, or any combination thereof. When implemented in software, they can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in this application are performed, in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, network equipment, user equipment, core network equipment, or other programmable devices.

[0106] The computer program or instructions may be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions may be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium may be any available medium that a computer can access, or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; or an optical medium, such as a digital video optical disc; or a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or non-volatile storage medium, or may include both volatile and non-volatile types of storage media.

[0107] The computer programs / instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper cables, fiber optic cables, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer-readable program instructions from the network and forwards those instructions to computer-readable storage media within the respective computing / processing device.

[0108] The computer program instructions used to perform the operations of this application may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​and conventional procedural programming languages. The computer-readable program instructions may be executed 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). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of this application.

[0109] Various aspects of this application are described herein with reference to flowchart illustrations and / or block diagrams according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0110] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0111] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0112] Each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the figures. For example, two consecutive blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that implementation in hardware, implementation in software, and implementation using a combination of software and hardware are equivalent.

[0113] This application provides a method, apparatus, device, medium, and product for intelligent agent interaction between applications. Its beneficial effects are as follows: by deploying system services supporting inter-agent interoperability protocols at the operating system layer, and providing an integration framework for these system services to several applications, the intelligent agents of the applications can access the system services based on the integration framework. After the intelligent agents access the system services, inter-application intelligent agent interaction is performed based on the forwarding of information between the intelligent agents by the system services. This solves the compatibility and maintenance problems of cross-application calls in the prior art, improves the determinism and efficiency of complex task execution between application intelligent agents, and enhances the user experience.

[0114] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications are also considered to be within the scope of protection of this application.

Claims

1. A method for intelligent agent interaction between applications, characterized in that, include: Deploy system services that support inter-agent interoperability protocols at the operating system layer; An integration framework for the system services is provided to several applications, enabling the intelligent agents of the applications to access the system services based on the integration framework; After the intelligent agent achieves the ability to access the system service, inter-application intelligent agent interaction is carried out based on the forwarding of information between the intelligent agents by the system service.

2. The inter-application agent interaction method as described in claim 1, characterized in that, The process of forwarding information between the agents based on the system service, and performing agent-to-application interaction, specifically includes: In response to a demand instruction from a demand-side agent, the system service sends a broadcast based on the demand instruction to at least one service-side agent, so that the service-side agent matching the demand instruction returns a status code. The system service sends the list of service-side intelligent agents corresponding to the status code to the demand-side intelligent agent, so that the demand-side intelligent agent can select the target intelligent agent from the list of service-side intelligent agents to execute the demand instruction; Based on the system service, a communication connection is established between the target agent and the demand agent, so that the demand agent can invoke the target agent to execute the demand instruction.

3. The inter-application agent interaction method as described in claim 2, characterized in that, In response to a demand instruction from a demand-side agent, the system service sends a broadcast based on the demand instruction to at least one service-side agent, causing service-side agents matching the demand instruction to return a status code, including: When a request instruction is received from a requesting agent, the system service sends a broadcast based on the request instruction to at least one service agent, so that the service agent receiving the broadcast can determine whether it has the ability to process the request instruction. If so, it returns a status code to the system service. The system service filters the status codes based on the response conditions in the demand instruction to determine the list of service agent intelligent agents.

4. The inter-application agent interaction method as described in claim 3, characterized in that, The response conditions include: the longest response time and / or the maximum number of responses; Wherein, if the response condition is the longest response time, the system service starts timing after receiving the demand instruction from the demand agent, and stops timing when the longest response time is reached. The system service determines the list of service agents based on the status codes received within the timing time. If the response condition is the maximum number of responses, the system service counts the received status codes. When the number of status codes is the maximum number of responses, the system service determines the list of service agent intelligent agents based on the received status codes.

5. The inter-application agent interaction method as described in claim 2, characterized in that, The system service sends the list of service-side agents corresponding to the status code to the request-side agent, so that the request-side agent can select the target agent from the list of service-side agents to execute the request instruction, including: The system service sends the list of service provider agents corresponding to the status code to the requesting agent, so that the requesting agent can select the target agent to execute the request instruction based on the list of service provider agents using a preset selection algorithm; wherein, the selection algorithm includes, but is not limited to: the fastest response matching algorithm, the optimal capability matching algorithm, or the random selection algorithm.

6. The inter-application agent interaction method as described in claim 2, characterized in that, The step of establishing a communication connection between the target agent and the demand agent based on the system service, so that the demand agent can invoke the target agent to execute the demand instruction, includes: A communication connection is established between the target intelligent agent and the demand-side intelligent agent based on the system service; When the system service receives information containing the target agent from the requesting agent, it forwards the information to the target agent so that the target agent can execute the request instruction.

7. The inter-application agent interaction method as described in claim 1, characterized in that, The method further includes: If the demand instruction from the demand-side intelligent agent is a long-running task, then the communication between intelligent agents is carried out through the system service via an asynchronous message interaction framework based on the message push mechanism.

8. The inter-application agent interaction method as described in claim 7, characterized in that, The asynchronous message interaction framework based on the message push mechanism enables communication between intelligent agents through the system service, including: After the demand-side intelligent agent and the service-side intelligent agent establish a communication connection based on the system service, the message push interface of the demand-side intelligent agent is provided to the service-side intelligent agent, so that after the service-side intelligent agent completes the demand instruction of the demand-side intelligent agent, it calls the message push interface to push the result to the demand-side intelligent agent.

9. The inter-application agent interaction method as described in claim 1, characterized in that, The integration framework that provides the system services is provided to several applications, enabling the intelligent agents of the applications to access the system services based on the integration framework, including: The system service integration framework is output to several applications in the form of a software development kit (SDK) so that the SDK can be integrated into the intelligent agents of several applications, enabling the intelligent agents of the applications to push and pull messages of the system service based on the SDK.

10. The inter-application agent interaction method as described in claim 1, characterized in that, The communication between the system service and the intelligent agent is implemented based on the Android system interface definition language AIDL.

11. The inter-application agent interaction method as described in claim 1, characterized in that, Before the inter-application agent interaction is performed by forwarding information between the agents based on the system service, the method further includes: In response to the binding service request of the intelligent agent, after the system service binds with the intelligent agent, it returns the invoked proxy object to the intelligent agent; wherein, the proxy object is used to connect to the service; In response to the agent's registration callback interface operation, the system service adds the callback interface to a thread-safe container to register a listener; wherein, the callback interface is used to allow the system service to send information to the agent; The system service sets up a death proxy for the callback interface to monitor the process of the agent. When the process of the agent terminates, the binding between the agent and the system service is released.

12. An intelligent agent interaction device between applications, characterized in that, include: The system service module is used to deploy system services that support inter-agent interoperability protocols at the operating system layer; A capability integration module is used to provide an integration framework for the system services to several applications, so that the intelligent agents of the applications can access the capabilities of the system services according to the integration framework. The information interaction module is used to perform inter-application agent interaction based on the forwarding of information between agents by the system service after the agent has achieved access to the system service.

13. The inter-application intelligent agent interaction device as described in claim 12, characterized in that, The process of forwarding information between the agents based on the system service, and performing agent-to-application interaction, specifically includes: In response to a demand instruction from a demand-side agent, the system service sends a broadcast based on the demand instruction to at least one service-side agent, so that the service-side agent matching the demand instruction returns a status code. The system service sends the list of service-side intelligent agents corresponding to the status code to the demand-side intelligent agent, so that the demand-side intelligent agent can select the target intelligent agent from the list of service-side intelligent agents to execute the demand instruction; Based on the system service, a communication connection is established between the target agent and the demand agent, so that the demand agent can invoke the target agent to execute the demand instruction.

14. The inter-application intelligent agent interaction device as described in claim 13, characterized in that, In response to a demand instruction from a demand-side agent, the system service sends a broadcast based on the demand instruction to at least one service-side agent, causing service-side agents matching the demand instruction to return a status code, including: When a request instruction is received from a requesting agent, the system service sends a broadcast based on the request instruction to at least one service agent, so that the service agent receiving the broadcast can determine whether it has the ability to process the request instruction. If so, it returns a status code to the system service. The system service filters the status codes based on the response conditions in the demand instruction to determine the list of service agent intelligent agents.

15. The inter-application intelligent agent interaction device as described in claim 14, characterized in that, The response conditions include: the longest response time and / or the maximum number of responses; Wherein, if the response condition is the longest response time, the system service starts timing after receiving the demand instruction from the demand agent, and stops timing when the longest response time is reached. The system service determines the list of service agents based on the status codes received within the timing time. If the response condition is the maximum number of responses, the system service counts the received status codes. When the number of status codes is the maximum number of responses, the system service determines the list of service agent intelligent agents based on the received status codes.

16. The inter-application intelligent agent interaction device as described in claim 13, characterized in that, The system service sends the list of service-side agents corresponding to the status code to the request-side agent, so that the request-side agent can select the target agent from the list of service-side agents to execute the request instruction, including: The system service sends the list of service provider agents corresponding to the status code to the requesting agent, so that the requesting agent can select the target agent to execute the request instruction based on the list of service provider agents using a preset selection algorithm; wherein, the selection algorithm includes, but is not limited to: the fastest response matching algorithm, the optimal capability matching algorithm, or the random selection algorithm.

17. The inter-application intelligent agent interaction device as described in claim 13, characterized in that, The step of establishing a communication connection between the target agent and the demand agent based on the system service, so that the demand agent can invoke the target agent to execute the demand instruction, includes: A communication connection is established between the target intelligent agent and the demand-side intelligent agent based on the system service; When the system service receives information containing the target agent from the requesting agent, it forwards the information to the target agent so that the target agent can execute the request instruction.

18. The inter-application intelligent agent interaction device as described in claim 12, characterized in that, The device is also used for: If the demand instruction from the demand-side intelligent agent is a long-running task, then the communication between intelligent agents is carried out through the system service via an asynchronous message interaction framework based on the message push mechanism.

19. The inter-application intelligent agent interaction device as described in claim 18, characterized in that, The asynchronous message interaction framework based on the message push mechanism enables communication between intelligent agents through the system service, including: After the demand-side intelligent agent and the service-side intelligent agent establish a communication connection based on the system service, the message push interface of the demand-side intelligent agent is provided to the service-side intelligent agent, so that after the service-side intelligent agent completes the demand instruction of the demand-side intelligent agent, it calls the message push interface to push the result to the demand-side intelligent agent.

20. The inter-application intelligent agent interaction device as described in claim 12, characterized in that, The integration framework that provides the system services is provided to several applications, enabling the intelligent agents of the applications to access the system services based on the integration framework, including: The system service integration framework is output to several applications in the form of a software development kit (SDK) so that the SDK can be integrated into the intelligent agents of several applications, enabling the intelligent agents of the applications to push and pull messages of the system service based on the SDK.

21. The inter-application intelligent agent interaction device as described in claim 12, characterized in that, The communication between the system service and the intelligent agent is implemented based on the Android system interface definition language AIDL.

22. The inter-application intelligent agent interaction device as described in claim 12, characterized in that, The device is also used for: In response to the binding service request of the intelligent agent, after the system service binds with the intelligent agent, it returns the invoked proxy object to the intelligent agent; wherein, the proxy object is used to connect to the service; In response to the agent's registration callback interface operation, the system service adds the callback interface to a thread-safe container to register a listener; wherein, the callback interface is used to allow the system service to send information to the agent; The system service sets up a death proxy for the callback interface to monitor the process of the agent. When the process of the agent terminates, the binding between the agent and the system service is released.

23. A computer device, characterized in that, It includes a processor and a memory, the memory storing a computer program configured to be executed by the processor, the processor executing the computer program to implement the inter-application agent interaction method as described in any one of claims 1 to 11.

24. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the device containing the computer-readable storage medium executes the computer program, it implements the inter-application intelligent agent interaction method as described in any one of claims 1 to 11.

25. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the inter-application agent interaction method as described in any one of claims 1 to 11.

Images