Networks are protected using many firewalls andencryption software’s. But many of them are not sufficient andeffective. Most intrusion detection systems for mobile ad hocnetworks are focusing on either routing protocols or itsefficiency, but it fails to address the security issues. Some of thenodes may be selfish, for example, by not forwarding the packetsto the destination, thereby saving the battery power. Someothers may act malicious by launching security attacks likedenial of service or hack the information. The ultimate goal ofthe security solutions for wireless networks is to providesecurity services, such as authentication, confidentiality,integrity,anonymity, and availability, to mobile users. Thispaper incorporates agents and data mining techniques toprevent anomaly intrusion in mobile adhoc networks. Homeagents present in each system collects the data from its ownsystem and using data mining techniques to observed the localanomalies. The Mobile agents monitoring the neighboring nodesand collect the information from neighboring home agents todetermine the correlation among the observed anomalouspatterns before it will send the data. This system was able to stopall of the successful attacks in an adhoc networks and reduce thefalse alarm positives.