Content processing method and system based on security protection agent

Abstract

The embodiment of the invention provides a content processing method based on a security protection agent, and the method comprises the steps: configuring an agent with a plurality of tools and front and rear hooks of the tools in a task flow, determining an information source of to-be-published content, loading the current environment information to generate a dynamic security protection rule, enabling the agent to execute the task flow, and enabling the security protection rule to be generated in the execution process. When the tool front hook is triggered, input information to be acquired by the subsequent tool is determined, the information input tool front hook provides the type of the subsequent tool and the input information to the created security protection thread, and the security protection thread calls a security protection rule; and judging whether the tool is allowed to transmit the content or not by using a security protection rule, the type of the tool and input information, and generating publishable content by the intelligent agent according to the content transmitted to the task flow terminal node. The intelligent agent is used for processing, the efficiency is high, and the publicity effect and the information safety can be improved by controlling the tool to output the publicity content through the dynamic safety protection rule.

Description

Technical Field

[0001] This application relates to the field of computers, and in particular to a content processing method and system based on a security-protected intelligent agent. Background Technology

[0002] In the field of content publishing, in order to improve the promotional effect, new products, new designs and new concepts are often incorporated into the promotional content. However, in order to improve the promotional effect while meeting the information security requirements, the content is often reviewed manually or through a pre-set review process before it is published. This method is insecure and inefficient.

[0003] In order to balance the effectiveness of publicity and information security, it is necessary to provide a new content processing method. Summary of the Invention

[0004] This specification provides a content processing method based on a security protection intelligent agent to improve the information security, publicity effect, and publishing efficiency of content publishing.

[0005] The method includes: A smart agent with multiple tools and hooks is configured in the task flow, the hooks including: a tool pre-hook and a tool post-hook; Determine the information source of the content to be published by the service recipient, load the current environmental information, and generate dynamic security protection rules based on the current environmental information; The agent creates a task flow thread to execute its task process. During the execution, when the tool pre-hook is triggered, it determines the input information that the subsequent tool needs to obtain. The input information includes at least one of the original information in the information source and the information processed by the model in the agent. The tool pre-hook provides the type of the subsequent tool and the input information to the created security protection thread. The security protection thread invokes the security protection rule, and uses the security protection rule, the type of the tool, and the input information to determine whether the tool is allowed to transmit content; The intelligent agent generates publishable content based on the content transmitted to the task flow terminal node.

[0006] Optionally, the step of determining whether to allow the tool to transmit content using the security protection rules, the type of the tool, and the input information includes: First, determine the type of the tool. If the tool is an internal transmission tool, then use an asynchronous execution program to execute the tool's instructions and the security protection rule's instructions in parallel to obtain the tool's processing result and the security protection's parallel judgment result. The parallel judgment result is whether the input information belongs to the current confidential content. After the tool is executed, it triggers the subsequent tool hook to call back the parallel judgment result, and controls whether the tool passes the processed content down based on the parallel judgment result. If the tool is an external transmission tool, then the following first and second programs are executed in parallel using an asynchronous execution program: The first program is to remove the corresponding tool hook, and the second program is to execute the instructions of the security protection rule and the instructions of the tool in sequence. During the sequence execution, after the instructions of the security protection rule generate the sequence judgment result of the security protection, the tool hook calls back the sequence judgment result and controls whether the tool is executed according to the sequence judgment result. After the tool hook is removed, the processed content is passed on after the tool is executed.

[0007] Optionally, generating dynamic security protection rules based on the current environmental information includes: Determine the time point of the last update of the security protection rules, determine the technical information submitted by the service provider on the authorized platform within the period from the time point to the present, and add the content of the technical information as non-confidential content to the security protection rules.

[0008] Optionally, the technical information includes: product design, interface design, and usage method description, etc.

[0009] Optionally, the intelligent agent is configured with a publishing document template, and the intelligent agent generates publishable content based on the content passed to the task flow terminal node, including: The intelligent agent processes the product's appearance, interface design, and usage method description, and fills them into the release file template to generate a product release file.

[0010] Optionally, the release document template has multiple sections, the information source includes the product design, interface design and usage method description, and the tools in the intelligent agent are configured to be associated with each section; The intelligent agent processes the product design, interface design, and usage method description, and populates them into the release file template to generate a product release file, including: Each tool delivers the processed content to at least one associated partition in the release file template.

[0011] Optionally, the method further includes: Tools that identify undelivered content will have their corresponding partition removed from the published file template.

[0012] This specification also provides an embodiment of a content processing system based on a security protection intelligent agent, including: The intelligent agent body module is configured to be an intelligent agent with multiple tools and hooks in the configuration task flow, the hooks including: tool pre-hook and tool post-hook; The protection module determines the information source of the content to be published by the service object, dynamically loads the current environmental information, and generates dynamic security protection rules based on the current environmental information. The agent ontology module creates a task flow thread to execute its task flow. When the tool pre-hook is triggered, it determines the input information that the tool needs to acquire. The input information includes at least one of the original information in the information source and the information processed by the model in the agent. The tool pre-hook provides the type of the subsequent tool and the input information to the security protection thread created by the protection module. The security protection thread created by the protection module calls the security protection rule and uses the security protection rule, the type of the tool, and the input information to determine whether the tool is allowed to transmit content. The intelligent agent ontology module generates publishable content based on the content transmitted to the task flow terminal node.

[0013] This specification also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps in any of the above embodiments. The various technical solutions provided in the embodiments of this specification, through configuring an intelligent agent with multiple tools and pre- and post-tool hooks in the task flow, determine the information source of the content to be published, load the current environmental information to generate dynamic security protection rules, and the intelligent agent executes its task flow. During execution, when the tool pre-hook is triggered, it determines the input information that the subsequent tool needs to obtain. The tool pre-hook provides the type of the subsequent tool and the input information to the created security protection thread. The security protection thread calls the security protection rules and uses the security protection rules, the tool type, and the input information to determine whether the tool is allowed to transmit content. The intelligent agent generates publishable content based on the content transmitted to the task flow terminal node. Using intelligent agents for processing is highly efficient, and controlling the output of publishable content by tools through dynamic security protection rules can improve the publicity effect and information security. Attached Figure Description

[0014] The accompanying drawings, which are provided to further illustrate this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application.

[0015] In the attached diagram: Figure 1 A schematic diagram illustrating the principle of a content processing method based on a security protection intelligent agent, provided for some embodiments of this specification; Figure 2 A schematic diagram illustrating the principle of a content processing method based on a security protection intelligent agent, provided for some embodiments of this specification; Figure 3 A schematic diagram of the structure of a content processing system based on a security protection intelligent agent, provided for some embodiments of this specification; Figure 4 This is a schematic diagram of a computer-readable medium provided for some embodiments of this specification. Detailed Implementation

[0016] The existing technology of setting fixed review processes in advance requires a lot of time and effort, and may even require cross-departmental cooperation between the operations department and the intellectual property department. It is less accurate, less efficient, and has high communication costs. It often requires reading the company's publicly available technical documents to determine whether certain content can be disclosed.

[0017] In other fields, with the development of artificial intelligence, the concept of intelligent agents has emerged. An intelligent agent is an abstract concept, and its specific form is a software system or an entity that carries this software. It can be regarded as a task process that can call multiple models and tools. It can analyze, judge and execute work by calling the models and tools designed for the task process. For example, a robot vacuum cleaner can generate a workflow based on its internal logic, and a human customer service intelligent agent can conduct dialogue.

[0018] Therefore, the applicant discovered that this convenience of intelligent agents can be used to process information to be published. However, the disclosure of technology often changes over time, and existing intelligent agents are mostly black box concepts that cannot be intervened. This makes it costly and time-delayed to construct new intelligent agents each time, and cannot truly meet the needs of information security and publicity.

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Figure 1 A schematic diagram illustrating the principle of a content processing method based on a security protection intelligent agent, provided for some embodiments of this specification, may include: S101: Configure an intelligent agent with multiple tools and hooks in the task flow, wherein the hooks include: a tool pre-hook and a tool post-hook; S102: Determine the information source of the content to be published by the service object, load the current environmental information, and generate dynamic security protection rules based on the current environmental information; S103: The agent creates a task flow thread to execute its task flow. During the execution, when the tool pre-hook is triggered, it determines the input information that the subsequent tool needs to obtain. The input information includes at least one of the original information in the information source and the information processed by the model in the agent. The tool pre-hook provides the type of the subsequent tool and the input information to the created security protection thread. S104: The security protection thread calls the security protection rule and uses the security protection rule, the type of the tool, and the input information to determine whether the tool is allowed to transmit content; S105: The intelligent agent generates publishable content based on the content transmitted to the task flow terminal node.

[0021] By configuring an agent with multiple tools and pre- and post-tool hooks in the task flow, the information source of the content to be published is determined. Current environmental information is loaded to generate dynamic security protection rules. The agent executes its task flow. During execution, when a tool's pre-hook is triggered, the input information to be obtained by the subsequent tool is determined. The tool's pre-hook provides the tool type and input information to the created security protection thread. The security protection thread invokes the security protection rules and uses the rules, tool type, and input information to determine whether the tool is allowed to transmit content. The agent generates publishable content based on the content transmitted to the task flow's terminal node. Using an agent for processing is highly efficient, and controlling the output of publicly available content through dynamic security protection rules improves both the effectiveness of publicity and information security.

[0022] The intelligent agent can have multiple models, and different models can be configured to process information sources of different data types, such as images, numbers, text, and graphics. It can call a large language model to process the information source, including text filtering and text concatenation, to generate fluent content. Third-party intelligent agent components can be used.

[0023] In the embodiments of this specification, hooks can be constructed before and after each tool to control the execution and output of the tool. A callback function can be configured in the hook to call back the judgment result of the security protection thread and give the judgment result of the security protection thread to control the execution and output of the tool. The hook before the tool is used to control whether the tool is executed, and the hook after the tool is used to control whether the tool outputs the processing result.

[0024] The tool pre-hook can be configured to be triggered to provide the type of the subsequent tool and the input information to the created security protection thread.

[0025] In this embodiment of the specification, determining whether to allow the tool to transmit content using the security protection rules, the type of the tool, and the input information includes: First, determine the type of the tool. If the tool is an internal transmission tool, then use an asynchronous execution program to execute the tool's instructions and the security protection rule's instructions in parallel to obtain the tool's processing result and the security protection's parallel judgment result. The parallel judgment result is whether the input information belongs to the current confidential content. After the tool is executed, it triggers the subsequent tool hook to call back the parallel judgment result, and controls whether the tool passes the processed content down based on the parallel judgment result. If the tool is an external transmission tool, then the following first and second programs are executed in parallel using an asynchronous execution program: The first program is to remove the corresponding tool hook, and the second program is to execute the instructions of the security protection rule and the instructions of the tool in sequence. During the sequence execution, after the instructions of the security protection rule generate the sequence judgment result of the security protection, the tool hook calls back the sequence judgment result and controls whether the tool is executed according to the sequence judgment result. After the tool hook is removed, the processed content is passed on after the tool is executed.

[0026] Figure 2 This diagram illustrates the principle of a content processing method based on a security protection intelligent agent, as provided in some embodiments of this specification. Figure 2 This illustrates one stage of an intelligent agent, namely, the judgment and subsequent processing based on the type of tool. An intelligent agent can be viewed as consisting of multiple such stages. Figure 2 middle: After the tool's pre-hook is triggered, the type of the subsequent tool and the input information are provided to the created security protection thread. The security protection thread (which can be embedded inside the agent) can determine whether the tool is allowed to execute first based on the tool type. If it is an internal transmission tool, it means that even if the tool processes content, it will not transmit the tool to the outside of the agent and will not leak information. Therefore, in order to improve processing efficiency, instructions from security protection rules can be used simultaneously. Figure 2 As shown in a1), it determines whether the input information is confidential, and then executes the tool's instructions and the hook's instructions (collectively referred to as...) in sequence. Figure 2As shown in a2), the tool processes the input information first without waiting for the judgment result of the security protection rule. These two processes are executed asynchronously in parallel, which is highly efficient. After the security protection rule obtains the parallel judgment result, the tool's hook is triggered, calling the concurrent judgment result. Based on the concurrent judgment result, the tool is controlled to pass the processing result. If the concurrent judgment result shows that the input information is confidential, it is not passed. Otherwise, it is passed to the next node in the task flow to continue processing. Of course, if the tool is the terminal node, it is passed to the file template. If the tool is an externally transmitting tool, it means that if input information is given to it, there is a risk that the tool will transmit the processing results to the intelligent body. Therefore, it is necessary to first determine whether the input information is confidential. If it is not confidential, then the tool can process it. Moreover, since the instructions of the security protection rules are executed first, followed by the instructions of the tool (the instructions of the security protection rules and the instructions of the tool are executed serially), as long as the tool executes, it means that the input information is not confidential. Therefore, the tool's post-hook is unnecessary and will only slow down the running time. Therefore, in the case of an externally transmitting tool, in addition to the serial execution of the instructions of the security protection rules and the tool's instructions, the removal of the tool's post-hook can be executed in parallel, while the serial execution of the instructions of the security protection rules and the tool's instructions (collectively referred to as...) Figure 2 As shown in b1), the hook after removing the tool (such as...) Figure 2 As shown in b2), these two are in a parallel relationship.

[0027] Specifically, if the instructions of the security protection rule are executed in parallel with the instructions of the tool, the judgment result of the instructions of the security protection rule is called the parallel judgment result; if the instructions of the security protection rule are executed serially with the instructions of the tool, the judgment result of the instructions of the security protection rule is called the serial judgment result.

[0028] In practice, it can be determined whether the memory containing the input data, process data, and output data of the tool during execution is externally transferable based on whether it can be read by other programs. If the data can be read by other programs, it is externally transferable memory; if it can only be read by the task flow and security protection rules of the intelligent agent, it is internally transferable memory.

[0029] In the embodiments of this specification, the information source of the content to be published by the service object may include the service object's internal files or project files on its server.

[0030] In one embodiment, the information source for determining the content to be published by the service recipient can be obtained from materials uploaded by the service provider. In this way, by having the service recipient actively upload materials, the intelligent agent automatically generates a customized publication document for them.

[0031] In order to maximize the disclosure of information while ensuring the security of trade secrets, thereby improving the publicity effect, recent content can be used as non-confidential content, which can further enhance the publicity effect.

[0032] Therefore, in this embodiment of the specification, generating dynamic security protection rules based on the current environmental information includes: Determine the time point of the last update of the security protection rules, determine the technical information submitted by the service provider on the authorized platform within the period from the time point to the present, and add the content of the technical information as non-confidential content to the security protection rules.

[0033] In terms of content, it can include the design patent application, patent content, etc.

[0034] Therefore, in the embodiments of this specification, the technical information includes: product design, interface design, and description of usage methods, etc.

[0035] In a practical application scenario, a user updates security protection rules on October 10th, submits a patent application for a new feature or design on October 20th, and needs to hold a product promotion event on October 30th. Although the patent application has not yet been published, it has already been filed, so even if it is used as promotional content, it will not affect the rights. Therefore, the new feature or design can be used as an information source to generate the publication document, which can promote the product to the greatest extent.

[0036] This method eliminates the need for repeatedly reviewing and manually checking which documents have been applied for, thus avoiding the risk of technical leaks due to human error.

[0037] In this embodiment of the specification, the intelligent agent is configured with a publishing file template, and the intelligent agent generates publishable content based on the content transmitted to the task flow terminal node, including: The intelligent agent processes the product's appearance, interface design, and usage method description, and fills them into the release file template to generate a product release file.

[0038] In this embodiment of the specification, the publishing file template has multiple partitions, the information source includes the product design, interface design and usage method description, and the tools in the intelligent agent are configured to be associated with each partition; The intelligent agent processes the product design, interface design, and usage method description, and populates them into the release file template to generate a product release file, including: Each tool delivers the processed content to at least one associated partition in the release file template.

[0039] After the content is transmitted, it can be rendered to obtain a displayable file, such as a slideshow.

[0040] In the embodiments described in this specification, the method further includes: Tools that identify undelivered content will have their corresponding partition removed from the published file template.

[0041] In the embodiments of this specification, the tool can perform various processing functions such as decomposing, filtering, and sorting input information, as well as data summarization, data summation, and chart generation. The specific functions of the tool are not the focus of this application and therefore will not be described in detail.

[0042] Among them, the security protection rules can record keywords for confidential and non-confidential content (content that can be published). In this way, by text matching, it can be determined which content can be published.

[0043] In the embodiments described in this specification, instructions for controlling parallel execution can be configured in the hook, or a unified parallel execution module can be configured outside the hook for unified scheduling, both of which are within the protection scope of this application.

[0044] Figure 3 A schematic diagram of a content processing system based on a security protection intelligent agent, provided for some embodiments of this specification, is shown. The system includes: The intelligent agent body module 301 is configured as an intelligent agent with multiple tools and hooks in the configuration task flow, the hooks including: tool pre-hook and tool post-hook; Protection module 302 determines the information source of the content to be published by the service object, dynamically loads the current environmental information, and generates dynamic security protection rules based on the current environmental information; The intelligent agent ontology module 301 creates a task flow thread to execute its task flow. When the tool pre-hook is triggered, it determines the input information that the tool needs to acquire. The input information includes at least one of the original information in the information source and the information processed by the model in the intelligent agent. The tool pre-hook provides the type of the subsequent tool and the input information to the security protection thread created by the protection module 302. The security protection thread created by the protection module 302 calls the security protection rule and uses the security protection rule, the type of the tool, and the input information to determine whether the tool is allowed to transmit content. The intelligent agent ontology module generates publishable content based on the content transmitted to the task flow terminal node.

[0045] The protection module 302 can be located in the server of the service provider, in the local processor where the agent is located, or embedded in the agent program as part of the agent platform, all of which should fall within the protection scope of this application.

[0046] Figure 3 The system shown can execute any of the methods described in the above method embodiments, which will not be elaborated here.

[0047] Figure 4 This is a schematic diagram of a computer-readable medium provided for embodiments of this specification.

[0048] accomplish Figure 1 The computer program for the method illustrated can be stored on one or more computer-readable media. The computer-readable storage medium may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. The program code for performing the operations of this invention can be written in any combination of one or more programming languages. The program code can be executed entirely on a user's computing device, partially on a user's computing device, as a standalone software package, partially on a user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be located via any type of network.

[0049] The above description is merely an embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.

Claims

1. A content processing method based on a security protection intelligent agent, characterized in that, include: A smart agent with multiple tools and hooks is configured in the task flow, the hooks including: a tool pre-hook and a tool post-hook; Determine the information source of the content to be published by the service recipient, load the current environmental information, and generate dynamic security protection rules based on the current environmental information; The agent creates a task flow thread to execute its task process. During the execution, when the tool pre-hook is triggered, it determines the input information that the subsequent tool needs to obtain. The input information includes at least one of the original information in the information source and the information processed by the model in the agent. The tool pre-hook provides the type of the subsequent tool and the input information to the created security protection thread. The security protection thread invokes the security protection rule, and uses the security protection rule, the type of the tool, and the input information to determine whether the tool is allowed to transmit content; The intelligent agent generates publishable content based on the content transmitted to the task flow terminal node.

2. The method according to claim 1, characterized in that, The step of determining whether to allow the tool to transmit content using the security protection rules, the tool type, and the input information includes: First, determine the type of the tool. If the tool is an internal transmission tool, then use an asynchronous execution program to execute the tool's instructions and the security protection rule's instructions in parallel to obtain the tool's processing result and the security protection's parallel judgment result. The parallel judgment result is whether the input information belongs to the current confidential content. After the tool is executed, it triggers the subsequent tool hook to call back the parallel judgment result, and controls whether the tool passes the processed content down based on the parallel judgment result. If the tool is an external transmission tool, then the following first and second programs are executed in parallel using an asynchronous execution program: The first program is to remove the corresponding tool hook, and the second program is to execute the instructions of the security protection rule and the instructions of the tool in sequence. During the sequence execution, after the instructions of the security protection rule generate the sequence judgment result of the security protection, the tool hook calls back the sequence judgment result and controls whether the tool is executed according to the sequence judgment result. After the tool hook is removed, the processed content is passed on after the tool is executed.

3. The method according to claim 2, characterized in that, The generation of dynamic security protection rules based on the current environmental information includes: Determine the time point of the last update of the security protection rules, determine the technical information submitted by the service provider on the authorized platform within the period from the time point to the present, and add the content of the technical information as non-confidential content to the security protection rules.

4. The method according to claims 1-3, characterized in that, The technical information includes: product design, interface design, and usage method description, etc.

5. The method according to claim 4, characterized in that, The intelligent agent is configured with a publishing file template. The intelligent agent generates publishable content based on the content passed to the task flow terminal node, including: The intelligent agent processes the product's appearance, interface design, and usage method description, and fills them into the release file template to generate a product release file.

6. The method according to claim 5, characterized in that, The release document template has multiple sections, the information source includes the product design, interface design and usage method description, and the tools in the intelligent agent are configured to be associated with each section; The intelligent agent processes the product design, interface design, and usage method description, and populates them into the release file template to generate a product release file, including: Each tool delivers the processed content to at least one associated partition in the release file template.

7. The method according to claim 6, characterized in that, Also includes: Tools that identify undelivered content will have their corresponding partition removed from the published file template.

8. A content processing system based on a security protection intelligent agent, characterized in that, include: The intelligent agent body module is configured to be an intelligent agent with multiple tools and hooks in the configuration task flow, the hooks including: tool pre-hook and tool post-hook; The protection module determines the information source of the content to be published by the service object, dynamically loads the current environmental information, and generates dynamic security protection rules based on the current environmental information. The agent ontology module creates a task flow thread to execute its task flow. When the tool pre-hook is triggered, it determines the input information that the tool needs to acquire. The input information includes at least one of the original information in the information source and the information processed by the model in the agent. The tool pre-hook provides the type of the subsequent tool and the input information to the security protection thread created by the protection module. The security protection thread created by the protection module calls the security protection rule and uses the security protection rule, the type of the tool, and the input information to determine whether the tool is allowed to transmit content. The intelligent agent ontology module generates publishable content based on the content transmitted to the task flow terminal node.

9. A computer-readable storage medium, characterized in that, The device stores a computer program that, when executed by a processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 7.

Images