Systems and methods for targeted attack protection using predictive sandboxing
Provided herein are systems and methods for targeted attack protection using predictive sandboxing. In exemplary embodiments, a method includes retrieving a Uniform Resource Locator (URL) from a message of a user and performing a preliminary determination to see if the URL can be discarded if it is not a candidate for sandboxing. The exemplary method includes computing a plurality of selection criteria factors for the URL if the URL passes the preliminary determination, each selection criteria factor having a respective factor threshold. The method can further include determining if any of the selection criteria factors for the URL exceeds the respective factor threshold for the respective selection criteria factor. Based on the determining, if any of the selection criteria factors exceeds the factor threshold for the selection criteria factor, the exemplary method includes automatically placing the URL in a sandbox for analysis.
1. A method for protecting users from malicious content by queuing one or more Uniform Resource Locator (URL) links pointing at linked content, the method comprising:
retrieving, by a hardware processor, a particular URL link pointing at linked content from an electronic message of a user;
tracking electronic messages that include the particular URL link, the tracking including determining a number of electronic messages that include the particular URL link;
determining if the number of electronic messages that include the particular URL link exceeds a predetermined message count threshold; and
selectively queuing the particular URL link for sandboxing for testing if determined that the number of electronic messages that include the particular URL link exceeds the predetermined message count threshold and the particular URL link has not been sandboxed for analysis in a configurable time period.
2. The method of claim 1 , wherein the determining the number of electronic messages that include the particular URL link comprises storing a message identifier for each of the electronic messages.
3. The method of claim 1 , further comprising:
normalizing the particular URL link to remove at least one of subdomains, query strings, and anchors, the determining the number of electronic messages that include the particular URL link being based on the normalized URL link.
4. The method of claim 3 , further comprising:
updating the number of electronic messages that include the particular URL link.
5. The method of claim 1 , wherein the selectively queuing the particular URL link for sandboxing comprises:
calculating a dynamic reputation score for the particular URL link;
comparing the dynamic reputation score to a predetermined reputation score threshold; and
selectively queuing the particular URL link for sandboxing if the dynamic reputation score exceeds the predetermined reputation score threshold.
6. The method of claim 5 , the selectively queuing the particular URL link for sandboxing further comprising:
calculating a targeted attack preparation score for the particular URL link;
comparing the targeted attack preparation score to a predetermined targeted attack preparation threshold; and
selectively queuing the particular URL link for sandboxing if the targeted attack preparation score exceeds the predetermined targeted attack preparation threshold.
7. The method of claim 1 , further comprising:
discarding the particular URL link from being sandboxed if a number of sandboxed URL links exceeds a predetermined sandbox limit.
8. The method of claim 1 , further comprising:
discarding the particular URL link from being sandboxed if a number of URL links that have been queued exceeds a domain count threshold, the domain count threshold being set per domain name.
9. The method of claim 8 , further comprising:
eliminating the domain count threshold if any sandboxed URL links are malicious.
10. The method of claim 9 , further comprising:
reinstituting the domain count threshold for the domain name after a set period of time or after a set number of URL links associated with the domain name are found to be non-malicious.
11. A system for protecting users from malicious content by queuing one or more Uniform Resource Locator (URL) links pointing at linked content, the system comprising:
at least one hardware processor; and
a memory for storing logic, the logic being executed by the at least one hardware processor to execute operations comprising:
retrieving a particular URL link pointing at linked content from an electronic message of a user;
tracking electronic messages that include the particular URL link, the tracking including determining a number of electronic messages that include the particular URL link;
determining if the number of electronic messages that include the particular URL link exceeds a predetermined message count threshold; and
selectively queuing the particular URL link for sandboxing for testing if determined that the number of electronic messages that include the particular URL link exceeds the predetermined message count threshold and the particular URL link has not been sandboxed for analysis in a configurable time period.
12. The system of claim 11 , wherein the determining the number of electronic messages that include the particular URL link comprises storing a message identifier for each of the electronic messages.
13. The system of claim 11 , the operations further comprising:
normalizing the particular URL link to remove at least one of subdomains, query strings, and anchors, the determining the number of electronic messages that include the particular URL link being based on the normalized URL link.
14. The system of claim 11 , the operations further comprising:
updating the number of electronic messages that include the particular URL link.
15. The system of claim 11 , wherein the selectively queuing the particular URL link for sandboxing comprises:
calculating a dynamic reputation score for the particular URL link;
comparing the dynamic reputation score to a predetermined reputation score threshold; and
selectively queuing the particular URL link for sandboxing if the dynamic reputation score exceeds the predetermined reputation score threshold.
16. The system of claim 15 , the selectively queuing the particular URL link for sandboxing further comprising:
calculating a targeted attack preparation score for the particular URL link;
comparing the targeted attack preparation score to a predetermined targeted attack preparation threshold; and
selectively queuing the particular URL link for sandboxing if the targeted attack preparation score exceeds the predetermined targeted attack preparation threshold.
17. The system of claim 11 , the operations further comprising:
discarding the particular URL link from being sandboxed if a number of sandboxed URL links exceeds a predetermined sandbox limit.
18. A non-transitory computer-readable storage medium having embodied thereon instructions, which, when executed by at least one hardware processor, perform steps of a method, the method comprising:
retrieving, by the at least one hardware processor, a particular Uniform Resource Locator (URL) link pointing at linked content from an electronic message of a user;
tracking electronic messages that include the particular URL link, the tracking including determining a number of electronic messages that include the particular URL link;
determining if the number of electronic messages that include the particular URL link exceeds a predetermined message count threshold; and
selectively queuing the particular URL link for sandboxing for testing if determined that the number of electronic messages that include the particular URL link exceeds the predetermined message count threshold and the particular URL link has not been sandboxed for analysis in a configurable time period.