Generating attack graphs in agile security platforms

1. A computer-implemented method for applying remediations for cyber security in enterprise networks, the method being executed by one or more processors and comprising:

providing, by a security platform, graph data defining a graph that is representative of an enterprise network, the graph comprising nodes and edges between nodes, the nodes comprising:

a set of configuration nodes each representing a condition of an asset of the enterprise network;

a set of rule nodes each representing an attack methodology; and

a set of impact nodes each representing an impact on the enterprise network,

wherein an edge between a configuration node and a rule node indicates that the condition represented by the configuration node enables the attack methodology represented by the rule node, and an edge between a rule node and an impact node indicates that the attack methodology represented by the rule node causes the impact;

identifying, by the security platform, one or more fixes for each of multiple respective conditions represented by configuration nodes of the graph;

incorporating, by the security platform, fix data representative of the identified one or more fixes into the graph, including linking each of the one or more fixes to the configuration node representing the respective condition;

defining, by the security platform and using the graph data, a set of fixes from the identified one or more fixes by executing an optimization problem for nullifying one or more impacts represented by impact nodes of the graph; and

scheduling, by the security platform, performance of each fix in the set of fixes.

2. The method of claim 1 , wherein each impact comprises an effect on an asset within the enterprise network.

3. The method of claim 1 , wherein each fix in the set of fixes has a cost associated therewith, wherein executing the optimization problem comprises optimizing the cost of the set of fixes.

4. The method of claim 3 , wherein the cost comprises one or more of a monetary cost, a temporal cost, and a resource cost.

5. The method of claim 1 , wherein the optimization problem is executed using a greedy algorithm and a coverage matrix.

6. The method of claim 5 , wherein the coverage matrix defines, for each fix of the one or more fixes, one or more impacts that the fix nullifies.

7. The method of claim 6 , wherein the coverage matrix further defines, for each fix of the one or more fixes, a linked impact with respect to one or more impacts affected by the fix.

8. The method of claim 1 , wherein scheduling, by the security platform, performance of each fix in the set of fixes is executed by performing a multi-objective optimization to minimize one or more of risk and cost for a given objective.

9. The method of claim 8 , wherein risk represents a risk of one or more vulnerabilities being exploited over time.

10. The method of claim 1 , wherein each fix is one of an official fix that wholly resolves a vulnerability of a configuration, a work-around fix that works around the vulnerability of the configuration, and a temporary fix that only temporarily resolves the vulnerability of the configuration.

11. The method of claim 1 , wherein assets comprise one or more of information technology (IT) assets and operational technology (OT) assets.

12. A non-transitory computer-readable storage medium coupled to one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations for applying remediations for cyber security in enterprise networks, the operations comprising:

providing, by a security platform, graph data defining a graph that is representative of an enterprise network, the graph comprising nodes and edges between nodes, each node representing an asset within the enterprise network, and each edge representing at least a portion of one or more lateral paths between assets in the enterprise network, the graph data comprising a set of configurations, each configuration affecting at least one impact;

determining, by the security platform, multiple sets of fixes for respective configurations;

incorporating, by the security platform, fix data representative of the multiple sets of fixes into the graph;

defining, by the security platform, a set of fixes comprising one or more fixes from the multiple sets of fixes by executing an optimization problem for nullifying one or more impacts, wherein the optimization problem is executed using a greedy algorithm and a coverage matrix that defines, for each fix in the multiple sets of fixes, one or more impacts that the fix nullifies; and

scheduling, by the security platform, performance of each fix in the set of fixes.

13. The non-transitory computer-readable storage medium of claim 12 , wherein each impact comprises an effect on an asset within the enterprise network.

14. The non-transitory computer-readable storage medium of claim 12 , wherein each fix in the set of fixes has a cost associated therewith, wherein executing the optimization problem comprises optimizing the cost of the set of fixes.

15. The non-transitory computer-readable storage medium of claim 14 , wherein the cost comprises one or more of a monetary cost, a temporal cost, and a resource cost.

16. The non-transitory computer-readable storage medium of claim 12 , wherein each fix in the set of fixes is associated with a respective configuration in the graph.

17. A system, comprising:

one or more computers; and

a computer-readable storage device coupled to the one or more computers and having instructions stored thereon which, when executed by the one or more computers, cause the one or more computers to perform operations for applying remediations for cyber security in enterprise networks, the operations comprising:

providing, by a security platform, graph data defining a graph that is representative of an enterprise network, the graph comprising nodes and edges between nodes, the nodes comprising:

a set of configuration nodes each representing a condition of an asset of the enterprise network;

a set of rule nodes each representing an attack methodology; and

a set of impact nodes each representing an impact on the enterprise network,

wherein an edge between a configuration node and a rule node indicates that the condition represented by the configuration node enables the attack methodology represented by the rule node, and an edge between a rule node and an impact node indicates that the attack methodology represented by the rule node causes the impact;

identifying, by the security platform, one or more fixes for each of multiple respective conditions represented by configuration nodes of the graph;

incorporating, by the security platform, fix data representative of the identified one or more fixes into the graph, including linking each of the one or more fixes to the configuration node representing the respective condition;

defining, by the security platform and using the graph data, a set of fixes from the identified one or more fixes by executing an optimization problem for nullifying one or more impacts represented by impact nodes of the graph; and

scheduling, by the security platform, performance of each fix in the set of fixes.

18. The system of claim 17 , wherein:

each fix in the set of fixes has a cost associated therewith,

executing the optimization problem comprises optimizing the cost of the set of fixes, and

the cost comprises one or more of a monetary cost, a temporal cost, and a resource cost.

19. The system of claim 17 , wherein:

the optimization problem is executed using a greedy algorithm and a coverage matrix,

the coverage matrix defines, for each fix of the one or more fixes, one or more impacts that the fix nullifies, and

the coverage matrix further defines, for each fix of the one or more fixes, a linked impact with respect to one or more impacts affected by the fix.

20. The system of claim 17 , wherein scheduling, by the security platform, performance of each fix in the set of fixes is executed by performing a multi-objective optimization to minimize one or more of risk and cost for a given objective, wherein risk represents a risk of one or more vulnerabilities being exploited over time.