A lot of privacy data are generated by using mobile social networking applications (MSNAs) and the values of user’s privacy data in those applications increase with the establishment and development of big data platform, which makes MSNAs the primary target to be analyzed. Therefore, it is important to analyze privacy leakage and protect user’s privacy in the MSNAs. However, the existing approaches of data leakage detection in the Android platform are not suitable for MSNAs, e.g. VetDroid are considered as an impractical means since they require users’ frequent participation; TaintDroid and the detection methods based on it require the modification of Android system or the modification and re-package of the application, so the cost of the experiment will increase, and the operating efficiency of the application will decrease apparently. In this paper, we propose a privacy leakage detection tool named X-Decaf (X-Posed based Detection of Cache File) as well as an auto-protection method named ATFed (Automatic Transparent File Encryption/Decryption) in MSNAs on the Android platform. These two methods are designed to solve the above-mentioned issues under the conditions of keeping low coupling with the Android system and posing low impacts on the original MSNA.