Multi-type vulnerability test set construction method and device based on template injection

Through template injection technology, the target program is constructed, which solves the problems of single type of vulnerability test set and single triggering method in the existing technology, and realizes accurate injection and detection of multiple types of vulnerabilities, improving the evaluation of vulnerability detection tools. accuracy.

CN118898074APending Publication Date: 2024-11-05Chinese People's Liberation Army Cyberspace Force Information Engineering University
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411086078.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-08
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

The existing vulnerability benchmark test set construction methods are mainly applicable to a single type of vulnerability test set. They cannot meet multiple types of scenarios for real vulnerabilities in software, and the triggering method is single, so they cannot apply to real vulnerability scenarios in open source software.

Method used

Using a multi-type vulnerability test set construction method based on template injection, filtering the objective function through the abstract syntax tree analysis of the target program, performing source code insertion and stain analysis, locate data flow introduction points and potential vulnerabilities, and convert preset Vulnerability templates are injected into it to form a test set of multiple vulnerability triggering methods.

Benefits of technology

A variety of vulnerability templates have been built, which can effectively inject potential vulnerabilities in the target program and accurately locate security vulnerabilities. The generated vulnerability test set is consistent with the real vulnerability characteristics. It is suitable for emerging vulnerability detection tools, and realizes the vulnerability detection tools. Fair assessment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118898074A_ABST
    Figure CN118898074A_ABST
Patent Text Reader

Abstract

The invention provides a multi-type vulnerability test set construction method and device based on template injection. The method comprises the steps of 1, analyzing an abstract syntax tree of a target program to screen out a target function; wherein the target function refers to a function which has a function body and can perform vulnerability injection; 2, source code instrumentation is carried out on the screened target function; 3, executing a target program containing source code instrumentation according to user input, and performing taint analysis to locate a data stream introduction point and a potential vulnerability point; and 4, injecting a preset vulnerability template between the data stream introduction point and the potential vulnerability point to form a vulnerability test set.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of fuzzy testing and evaluation technology, and in particular to a method and device for constructing a multi-type vulnerability test set based on template injection. Background Art

[0002] Most of the current vulnerability benchmark test set construction methods are only applicable to the construction of a single type of vulnerability test set. The vulnerability types included are single and do not conform to the multi-type scenarios of real vulnerabilities in software. In addition, the vulnerability triggering method is single and cannot be applied to the triggering scenarios of real vulnerabilities in open source software. Summary of the invention

[0003] The present invention provides a method and device for constructing a multi-type vulnerability test set based on template injection, which aims to generate a vulnerability template containing multiple types of vulnerabilities in multiple types of scenarios that conform to real vulnerabilities, and by embedding the vulnerability template in potential vulnerability points in the target program, the generated vulnerability test set has multiple vulnerability triggering modes, which can simulate real software vulnerabilities.

[0004] On the one hand, the present invention provides a method for constructing a multi-type vulnerability test set based on template injection, comprising:

[0005] Step 1: Analyze the abstract syntax tree of the target program to filter out the target function; wherein the target function refers to a function that has a function body and can perform vulnerability injection;

[0006] Step 2: Instrument the source code of the selected target function;

[0007] Step 3: Execute the target program with source code instrumentation according to user input, and perform taint analysis to locate data flow introduction points and potential vulnerability points;

[0008] Step 4: Inject the preset vulnerability template between the data flow introduction point and the potential vulnerability point to form a vulnerability test set.

[0009] Furthermore, step 2 specifically includes:

[0010] A probe statement is embedded in the target function; wherein the probe statement has a callback function embedded in it and carries the sequence number of the probe statement and the code location information of the probe statement.

[0011] Furthermore, step 3 specifically includes:

[0012] Taint each byte of user input and track the propagation path of user input in the target program;

[0013] Traverse the abstract syntax tree of the target program and determine the data flow introduction points and potential vulnerability points based on the information and taint markers carried by the probe statements.

[0014] Furthermore, the ASTMatcher tool is used to traverse the abstract syntax tree of the target program.

[0015] Furthermore, the preset vulnerability template includes vulnerability type, preconditions, free variables and vulnerability code; wherein the vulnerability type is used to indicate the vulnerability category represented by the vulnerability code; the preconditions are used to indicate the conditions that need to be met when the vulnerability code is injected; and the free variables represent preset variables in the vulnerability code.

[0016] On the other hand, the present invention provides a device for constructing a multi-type vulnerability test set based on template injection, comprising:

[0017] A function screening module, used to analyze the abstract syntax tree of the target program to screen out a target function; wherein the target function refers to a function having a function body and capable of vulnerability injection;

[0018] The instrumentation module is used to instrument the source code of the selected target functions;

[0019] The taint analysis module is used to execute the target program containing source code instrumentation according to user input and perform taint analysis to locate data flow introduction points and potential vulnerability points;

[0020] The vulnerability injection module is used to inject the preset vulnerability template between the data flow introduction point and the potential vulnerability point to form a vulnerability test set.

[0021] Beneficial effects of the present invention:

[0022] (1) The present invention constructs a series of vulnerability templates for various types of vulnerabilities by deeply analyzing the contextual environments of various types of known vulnerabilities. These vulnerability templates can be effectively injected into potential vulnerability points in the target program, thereby solving the problem that the current vulnerability test set contains a single type of vulnerability and does not conform to the multi-type scenarios of real vulnerabilities in the software.

[0023] (2) The present invention firstly performs source code instrumentation and abstract syntax tree matching technology on the target program, so as to accurately locate the structural position where the security vulnerability may exist in the source code corresponding to the target program; then, taint analysis is used to further screen the above structural positions and confirm the effective data flow introduction points and potential vulnerability points; finally, the constructed vulnerability template is injected between the effective data flow introduction points and the potential vulnerability points to complete the vulnerability injection into the open source software; in this process, the mutual cooperation of source code instrumentation, abstract syntax tree matching and taint analysis enables the present invention to automatically identify and extract different types of vulnerability point information in the target program.

[0024] (3) The present invention designs and constructs a series of vulnerability templates based on the specific features and context information of various known types of vulnerabilities. These vulnerability templates not only contain vulnerability codes, but also provide the prerequisites that need to be met to inject the vulnerability codes, the vulnerability categories represented by the vulnerability codes, and other information, thus providing a basis for the precise injection of vulnerabilities.

[0025] (4) The present invention is applicable to emerging vulnerability detection tools, especially in fuzzy test evaluation. Since the vulnerability test set generated by the present invention is more consistent with the characteristics of real vulnerabilities and has multiple vulnerability triggering modes, the vulnerability test set created by the present invention can achieve fair evaluation of vulnerability detection tools. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A schematic diagram of a flow chart of a method for constructing a multi-type vulnerability test set based on template injection provided by an embodiment of the present invention;

[0027] Figure 2 Example code provided for embodiments of the present invention;

[0028] Figure 3 The embodiment of the present invention provides Figure 2 Abstract syntax tree for the code shown;

[0029] Figure 4 A schematic diagram of a process of using an abstract syntax tree to match an array to access x[i] or a pointer to dereference *p provided in an embodiment of the present invention;

[0030] Figure 5 A schematic diagram of embedding a probe statement at the beginning of each line of code in a target function provided by an embodiment of the present invention;

[0031] Figure 6 A schematic diagram of a vulnerability template provided by an embodiment of the present invention;

[0032] Figure 7 A schematic diagram of injecting a vulnerability template into file software provided by an embodiment of the present invention;

[0033] Figure 8 A schematic diagram of a vulnerability sample provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] like Figure 1 As shown, an embodiment of the present invention provides a method for constructing a multi-type vulnerability test set based on template injection, comprising the following steps:

[0036] S101: Analyze the abstract syntax tree of the target program to filter out a target function; wherein the target function refers to a function having a function body and capable of vulnerability injection;

[0037] It is understandable that the target function cannot be a system function, because although system functions (such as printf) may have theoretical vulnerabilities, their source code cannot be modified (that is, there cannot be potential vulnerability injection points or data flow introduction paths in system functions), so vulnerability injection cannot be performed. The target function cannot be a function or structure that only has a declaration but no specific implementation, because the absence of an actual function body means that vulnerability injection cannot be performed.

[0038] Specifically, the abstract syntax tree records key information such as node type, node parameters, and the relationship between nodes in detail, so function screening can be implemented based on the abstract syntax tree of the target program's source code. Figure 2 The example code and its corresponding abstract syntax tree shown in Figure 3 ) is a specific example of this process.

[0039] Specifically, when performing function screening, the focus of the analysis includes elements such as system call functions, structures, and function declarations. These elements are often irrelevant to the subsequent in-depth analysis and should be excluded during the analysis process. The purpose of the analysis is to identify those functions that have actual function bodies in the source code and include these functions in the "whitelist" for analysis and processing in subsequent steps. For example, Figure 2 As shown in , the setparam function should be added to the whitelist because it has an actual function body; while the printf function, as a system call, should be filtered out in the process.

[0040] S102: Perform source code stubbing on the selected target function;

[0041] It can be understood that this step uses source code stubbing to effectively track and record the specific location of program execution, data flow introduction point and other key information during the taint analysis phase.

[0042] Specifically, source code instrumentation is implemented by embedding probe statements inside whitelist functions. Each probe statement has a callback function embedded inside it. These callback functions are used specifically for detailed analysis of taint data during the taint analysis phase. The parameter information carried by the probe statement mainly includes the serial number and the corresponding code location. The main function of the probe statement is to accurately mark the specific execution location of the code during the taint analysis process. By analyzing the information in the probe statement, the execution path of the current program and its specific location can be determined. Figure 5 As shown, the probe statement not only participates in the entire process of taint analysis, but also provides information about the key code locations during program execution.

[0043] Specifically, the embedding position of the probe statement can be selected as needed. The source code plugging in this step performs lexical and syntactic analysis on the source program file without executing the program itself, thereby traversing the abstract syntax tree (AST) based on these analyses, accurately locating the position suitable for plugging, and implanting predefined specific plugging statements in these selected positions. The lexical and syntactic analysis in this process ensures the high accuracy and clear targeting of the entire plugging operation, making the modification of the source file both accurate and purposeful. Source code plugging is completed before the program is executed, which means that it has little impact on the performance of the program at runtime, and is limited to the time required for the execution of the inserted function. Therefore, the application of source code plugging technology effectively reduces the additional overhead when the program is running, and at the same time provides an efficient and accurate method for testing and analyzing the program.

[0044] As an implementable method, this embodiment inserts the probe statement at the beginning of each line of code in the whitelist function by default.

[0045] S103: executing the target program containing the source code stub according to the user input, and performing taint analysis to locate the data flow introduction point and potential vulnerability points;

[0046] It can be understood that this step mainly uses taint analysis to parse the propagation path of user input data in the target program. This involves identifying the user input offset position corresponding to the free variable in the target program, converting the user input data into data with taint marks, and applying taint labels to these data. This configuration method of taint labels plays a key role in the propagation process of tainted data and the efficiency of taint status analysis.

[0047] Specifically, in order to guide the data flow from user input to potential vulnerability points and optimize the speed of taint analysis, the embodiment of the present invention does not use bit-level taint label setting, but chooses a more efficient byte-level taint label configuration, that is, taint marking each byte of user input. By tracing the propagation path of user input in the target program, the relationship between the free variables of the target program and the user input is further analyzed, and the connection between these free variables and potential vulnerability injection points is revealed.

[0048] Specifically, by using the probe statements inserted in the source code and the byte-level taint analysis technology, we can accurately locate the potential vulnerability injection points and the specific paths of the user input through the data flow introduction point to reach the location of these potential vulnerability injection points. By analyzing and storing the location information of the vulnerability injection points in the entire program and the user input offset location information corresponding to the free variables, we provide key information for selecting vulnerability injection points for vulnerability template injection and writing PoC.

[0049] In addition, in order to ensure that the taint source can be accurately propagated during the taint analysis process, the embodiment of the present invention also formulates taint propagation rules. These rules cover the marking logic of function operands and target numbers in taint analysis under different types of instructions, and the details are listed in Table 1. For example, Tag(op1) is used to indicate whether the operand at address op1 is tainted. For example, Tag(PSW)←Tag(op1)||Tag(op2) means that once any operand of op1 or op2 is marked as tainted, the flag register (PSW) will also be marked accordingly to achieve effective propagation of taint.

[0050] It is worth noting that when performing stack operations (including popping and pushing), although the rules do not explicitly specify the target operands for pushing and the source operands for popping, the relevant operands can still be accurately identified and obtained by analyzing the memory environment when pushing. This set of rules is designed to enhance the efficiency and accuracy of the taint analysis process by precisely controlling the propagation of taints.

[0051] Table 1 Taint propagation rules

[0052]

[0053] By analyzing the information in the probe statement, the execution path of the current program and its specific location can be determined. In actual applications, the DWARF debugging information of the target program can also be used to obtain the local variables on the execution path and their specific locations in memory or registers, so as to accurately perform taint tracking analysis.

[0054] Specifically, the embodiment of the present invention analyzes the structural framework of the program by deeply traversing the abstract syntax tree, thereby accurately locating the position of free variables and identifying potential vulnerability points.

[0055] It is understandable that, in general, the target program usually contains tens of thousands of lines of code, and the corresponding abstract syntax tree therefore contains a large amount of information. The abstract syntax tree AST is composed of multiple types of nodes (as shown in Table 2), each of which is an instance of a declaration (decl) or statement (stmt) class, so corresponding processing measures need to be taken for different types of nodes during the traversal process. In order to optimize the analysis efficiency of AST, the present invention introduces pattern matching technology to accurately find specific code fragments.

[0056] Table 2 Abstract syntax tree nodes

[0057] AST Node illustrate AST Node FunctionDecl Function root node FunctionDecl CompoundDecl Function body identification CompoundDecl ParamVarDecl Parameter Identification ParamVarDecl VarDecl Local and global variable declaration identifiers VarDecl CompoundStmt Code block identifier CompoundStmt DeclStmt Local variable declaration DeclStmt IfStmt Conditional statement identifier IfStmt

[0058] In response to the system analysis requirements, the embodiment of the present invention refines and extracts several key pattern types, such as variable declaration (vardecl), function declaration (funcdecl), type declaration (typedecl) and function call expression (callexpr). By writing special matching functions for these patterns, the present invention can more accurately identify the code location required for analysis in the process of traversing AST. This method based on pattern type extraction greatly improves the ability to perform effective analysis in complex AST structures, allowing the present embodiment to more accurately locate the analysis target.

[0059] Specifically, the embodiment of the present invention makes full use of the ASTMatcher tool provided by clang, which greatly simplifies the matching process of abstract syntax tree nodes. ASTMatcher enables users to accurately match AST nodes directly in the form of templates by defining a series of matching rules. This process not only helps to identify nodes, but also allows users to access key information such as their attributes and source code location through the C++ interface of the node, greatly enhancing the ability to analyze the source code structure.

[0060] For example, this example shows how to traverse the AST to match instances of array access x[i] or pointer dereference *p in the source code. The specific matching rules and examples are in Figure 4 The application of this method effectively promotes the depth and breadth of source code analysis and provides support for identifying specific code patterns and structures.

[0061] S104: Injecting a preset vulnerability template between the data flow introduction point and the potential vulnerability point to form a vulnerability test set.

[0062] Specifically, this embodiment builds a set of multiple vulnerability templates based on the CWE vulnerability library in the Juliet C / C++ test set in NIST (National Institute of Standards and Technology) (specifically, there are 11 different vulnerability type templates, covering common security vulnerability types, such as buffer overflow, null pointer dereference, integer overflow, etc.), which are key factors for achieving effective vulnerability injection. The vulnerability template mainly includes four core components: vulnerability type, precondition, free variable and vulnerability code.

[0063] Among them, the vulnerability type is used to indicate the vulnerability category represented by the vulnerability code. This information is crucial for selecting a suitable vulnerability template. The preconditions specify the specific conditions that the vulnerability code must meet when injecting the vulnerability, such as the context of the vulnerability injection location, such as the variable length. These conditions directly affect whether the vulnerability template can be successfully injected at a specific code location. By comparing the preconditions with the context of the target program, it is analyzed whether the conditions described therein are met, so as to determine whether it is suitable to inject the vulnerability at that location. By setting the preconditions, it can be ensured that the preconditions match the abstract syntax tree of the target program, and the code segment suitable for injecting the vulnerability can be accurately located. The free variables contain a list of variables preset in the vulnerability code. By matching the free variables in the vulnerability template with the relevant variables in the target program, it is ensured that the vulnerability code can be correctly integrated into the target program to achieve effective injection of the vulnerability. The vulnerability code describes the specific code to be inserted into the target program. This part constitutes the essence of the vulnerability template and contains all the necessary code to implement vulnerability injection.

[0064] Specifically, the vulnerability injection process relies on the code replacement function provided by the Clang tool, that is, the replacement method in the tool is used to insert the vulnerability code in the vulnerability template into the selected source code location. In the source code modification stage, the selected vulnerability code is implanted into the corresponding position of the target program, and compared with the vulnerability template conditions, the relevant variables in the scope of the target program are searched and matched, and then the preset variables in the vulnerability template are replaced, thereby completing the vulnerability injection.

[0065] Taking the CWE-122 vulnerability as an example, the vulnerability template explicitly requires the existence of a free variable named src, which is marked as a tainted variable in the analysis results, and the length of this tainted variable must exceed 16 characters, as a prerequisite for vulnerability injection. In this way, the vulnerability template provides effective guidance for finding suitable vulnerability injection points in the source code.

[0066] In the embodiment of the present invention, the process of extracting vulnerability templates from specific vulnerability instances is discussed in detail, and the heap overflow vulnerability in the CWE-122 type is used as an example for analysis. Figure 8 As shown in the figure, it comes from a program fragment, which reveals a critical security vulnerability: when using the memcpy() function to copy memory, its target operand points to a member variable in the structure, but the copy length is incorrectly set to the length of the entire structure. This will cause the source data length of the copy operation to exceed the target buffer range, causing a heap overflow, which will trigger the vulnerability when the program is running. This vulnerability is widely present in actual software and is easily detected by security tools such as sanitizers.

[0067] For the above vulnerability codes, we have constructed corresponding vulnerability templates by deeply analyzing and extracting key features, such as Figure 6 This template aims to introduce the previously described heap overflow vulnerability by injecting specific memcpy function call code into the target program. During the template injection process, in order to ensure the high compatibility and execution correctness of the vulnerability code with the target program, the predefined variables in the template will be replaced with the variables that actually exist in the target program.

[0068] Further, Figure 7 It shows that after applying the vulnerability template in the file software, the source operand buf of the memcpy function call is successfully replaced with the original variable of the same name in the program, which proves the flexibility and adaptability of the vulnerability injection technology. Importantly, the buf variable involved is not a newly generated variable during the vulnerability injection process, but a pre-existing variable in the software itself, which further highlights the ability of the vulnerability injection method to integrate existing code structures and logical structures in practical applications. In summary, through a series of in-depth analyses, not only the vulnerability template was successfully extracted from the specific vulnerability instance, but also the injection process of the vulnerability template in the real software environment was demonstrated.

[0069] It can be understood that due to the wide application of tainted variables and the diversity of vulnerability templates in the Juliet C / C++ test set, the vulnerability test set construction method proposed in the present invention has the potential to generate comprehensive and diverse vulnerable programs.

[0070] Corresponding to the above method, an embodiment of the present invention further provides a device for constructing a multi-type vulnerability test set based on template injection, including a function screening module, a stub module, a taint analysis module and a vulnerability injection module.

[0071] Specifically, the function screening module is used to analyze the abstract syntax tree of the target program to screen out the target function; wherein, the target function refers to a function with a function body and capable of vulnerability injection. The stub module is used to perform source code stubs on the screened target function; the taint analysis module is used to execute the target program with source code stubs according to user input, and perform taint analysis to locate the data flow introduction point and potential vulnerability points; the vulnerability injection module is used to inject the preset vulnerability template between the data flow introduction point and the potential vulnerability point to form a vulnerability test set.

[0072] It should be noted that the device for constructing a multi-type vulnerability test set based on template injection provided in the embodiment of the present invention is for the purpose of implementing the above method. Its specific functions can be referred to the above method embodiment and will not be described in detail here.

[0073] The present invention proposes a method for evaluating the effectiveness of vulnerability analysis tools. By constructing a vulnerability test set containing multiple real vulnerability features, it can be effectively applied to the current mainstream framework, thereby improving the accuracy and comprehensiveness of the evaluation framework.

[0074] The present invention not only injects vulnerability code into the program, but also ensures that the injected code has multiple real vulnerability features through a carefully designed injection process. This injection method is more in line with the actual vulnerability scenario, making the evaluation result more reliable and practical.

[0075] The vulnerability test set constructed by the present invention covers 11 different types of vulnerability codes, which means that the vulnerability test set has high diversity and coverage, and can more comprehensively evaluate the performance and accuracy of vulnerability analysis tools.

[0076] In summary, the present invention is not only applicable to the evaluation field of vulnerability analysis tools, but also has a driving effect on the entire test set evaluation ecosystem. By providing more accurate and comprehensive test sets, it promotes the development of the vulnerability analysis tool evaluation field and provides more reliable security protection for related industries.

[0077] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a multi-type vulnerability test set based on template injection, characterized in that: include: Step 1: Analyze the abstract syntax tree of the target program to filter out the target function; wherein the target function refers to a function that has a function body and can perform vulnerability injection; Step 2: Instrument the source code of the selected target function; Step 3: Execute the target program with source code instrumentation according to user input, and perform taint analysis to locate data flow introduction points and potential vulnerability points; Step 4: Inject the preset vulnerability template between the data flow introduction point and the potential vulnerability point to form a vulnerability test set; wherein the preset vulnerability template is constructed based on the context environment of multiple types of known vulnerabilities.

2. According to the method for constructing a multi-type vulnerability test set based on template injection according to claim 1, it is characterized in that: Step 2 specifically includes: A probe statement is embedded in the target function; wherein the probe statement has a callback function embedded in it and carries the sequence number of the probe statement and the code location information of the probe statement.

3. According to the method for constructing a multi-type vulnerability test set based on template injection according to claim 2, it is characterized in that: Step 3 specifically includes: Taint each byte of user input and track the propagation path of user input in the target program; Traverse the abstract syntax tree of the target program and determine the data flow introduction points and potential vulnerability points based on the information and taint markers carried by the probe statements.

4. According to the method for constructing a multi-type vulnerability test set based on template injection according to claim 3, it is characterized in that: Use the ASTMatcher tool to traverse the abstract syntax tree of the target program.

5. According to the method for constructing a multi-type vulnerability test set based on template injection according to claim 1, it is characterized in that: The preset vulnerability template includes vulnerability type, precondition, free variable and vulnerability code; wherein, vulnerability type is used to indicate the vulnerability category represented by the vulnerability code; precondition is used to indicate the conditions that need to be met when the vulnerability code is injected; free variable represents the preset variable in the vulnerability code.

6. A device for constructing a multi-type vulnerability test set based on template injection, characterized in that: include: A function screening module, used to analyze the abstract syntax tree of the target program to screen out a target function; wherein the target function refers to a function having a function body and capable of vulnerability injection; The instrumentation module is used to instrument the source code of the selected target functions; The taint analysis module is used to execute the target program containing source code instrumentation according to user input and perform taint analysis to locate data flow introduction points and potential vulnerability points; The vulnerability injection module is used to inject a preset vulnerability template between the data flow introduction point and the potential vulnerability point to form a vulnerability test set; wherein the preset vulnerability template is constructed based on the context environment of multiple known types of vulnerabilities.

Citation Information

Cited By

  • Methods and systems for generating customizable test cases from resources with embedded vulnerability indicators

    US20260300499A1