Wireless Sensor Networks (WSNs) consist of sensor nodes deployed in a manner to collect information about surrounding environment. In this paper design collaborative multi agent trust-based data mining technique for intrusion detection in wireless sensor network has been proposed .In their distributed nature, multi hopdata forwarding, and open wireless medium are the factors that make WSNs highly vulnerable to security attacks at various levels. Intrusion Detection Systems (IDSs) can play an important role in detecting and preventing security attacks. This paper presents current Intrusion Detection Systems and some open research problems related to WSN security. In order to achieve both a higher detection rate and a lower false positive rate of internal node intrusion detection in layer-cluster wireless sensor networks, an intrusion detection scheme based on the use of both a multi-agent system and a node trust value is proposed. In this scheme, the multi-agent model framework is established in both the cluster heads and the ordinary sensor nodes to perform intrusion detection. First, various, typical node trust attributes are defined and Mahalanob is distance theory is used to judge whether these attributes are normal. Second, the node trust value is calculated and updated based on the combination of the Beta distribution and a tolerance factor. Finally, node intrusion detection is realized. Simulation results demonstrate that the modified scheme has a higher detection rate and a lower false positive rate, even when several types of intrusions are present.