Granted Patent B2

US 8,490,196 · App. 12/851,516 · Granted Jul 16, 2013

Abstract

Claims (36)

1. A method for extending automated penetration testing of a target network comprising, the steps of:

a. computing a scenario, wherein the step of computing the scenario comprises the steps of

i. translating a workspace having at least one target computer in the target network, to a planning definition language;

ii. translating penetration modules available in a penetration testing framework to the planning definition language; and

iii. defining a goal in the target network and translating the goal into the planning definition language;

b. building a knowledge database with information regarding the target network, properties of hosts in the network, parameters and running history of modules in the penetration testing framework; and

c. running an attack plan solver module, wherein the attack plan solver module performs the steps of:

i. running an attack planner using the scenario as input, to produce at least one attack plan that achieves the goal; and

ii. executing actions defined in the at least one attack plan against the target network from the penetration testing framework.

2. The method of claim 1 wherein, prior to running the attack plan solver module, a pre-planner processor module is run using the scenario as input to produce a new scenario, wherein the pre-planner processor module parses the scenario to remove actions that cannot be executed against the hosts described in the scenario.

3. The method of claim 1 wherein, prior to running the attack plan solver module, a pre-planner processor module is run using the scenario as input to produce a new scenario, wherein the pre-planner processor module parses the scenario to combine into a single asset all those assets that can be targeted by the same actions and pose the same preconditions for these actions.

4. The method of claim 1 wherein, prior to running the attack plan solver module, a pre-planner processor module is run using the scenario as input to produce a new scenario, wherein the pre-planner processor module parses the scenario to combine into a single action all those actions that can be used to compromise a previously specified asset.

5. The method of claim 1 wherein the attack planner is selected from the group consisting of a proactive planner, a reactive planner, and a custom planner.

6. The method of claim 1 , where the attack planner solver module, after computing the attack plan, executes a post-planner processor module, wherein the post-planner processor module runs an iterative process comprising the steps of:

i. executing the first action in the plan;

ii. monitoring the result of the first action in the workspace to ascertain if the effect predicted by the attack plan was successful; and

iii. in case the action failed, modifying the scenario to produce a modified scenario, and calling the attack planner to re-calculate a modified plan with the modified scenario.

7. The method of claim 1 wherein an end user defines the goal to be achieved by the system.

8. The method of claim 1 , wherein the actions defined in the attack plans by the attack plan solver module are run in parallel when the preconditions for these actions are satisfied and as limited by the penetration testing framework.

9. The method of claim 1 , wherein the attack plan solver module updates the knowledge base after each action is executed.

10. The method of claim 1 , wherein the planning definition language is a planning domain definition language.

11. The method of claim 1 wherein the attack plan softer module comprises a pre-planner processor module that uses information available in the knowledge database and the scenario, which is in the planning definition language, to modify the scenario.

12. The method of claim 11 wherein the attack plan solver module further comprises a planner processor module and a post-planner processor module, the method further comprising:

sending the modified scenario from the pre-planner processor module to the planner processor module,

computing a single plan at the planner processor module, and

calling from the post-planner processor module a plan translator and executing each step of the single plan using the penetration testing framework.

13. The method of claim 11 wherein the attack plan solver module further comprises a planner processor module and a post-planner processor module, the method further comprising:

sending the modified scenario from the pre-planner processor module to the planner processor module;

computing different plans at the planner processor module; and

executing the different plans in parallel using the penetration testing framework until the goal is reached by one of the plans.

14. The method of claim 11 wherein the attack plan solver module further comprises a planner processor module and a post-planner processor module, the method further comprising:

sending the modified scenario from the pre-planner processor module to the planner processor module;

computing a first plan with the planner processor module;

forwarding the first plan to the post-planner processor module, which starts executing each step of the first plan by calling a plan translator to translate the first plan into commands of the penetration testing framework; and

monitoring with the planner processor module when new information is updated in the scenario.

15. The method of claim 1 wherein the computed scenario is in the planning definition language and wherein running the attack planner using the scenario as input comprises using the computed scenario in the planning definition language to produce at least one attack plan that achieves the goal.

Assignments (20)