Privacy Violation: Android Insecure Indexing

Creates a search index of private or sensitive documents, which can compromise user privacy.

Insecure indexing threatens the data confidentiality of the websites and mobile applications that enable indexing for efficient full-text search; including private or confidential data in documents indexed by Android search engine which it can result in information leakage about the existence and content of the data. In the process of indexing, information is collected and stored. An attacker can later retrieve this information (although not trivially), usually through a series of queries to the search facility on an Android mobile device.

AppSearch library, a high-performance Android on-device search solution, is a typical example of data indexing, particularly when a certain application contributes to a system-wide central index on the device, which is called platform storage. Data access within the central index is restricted to data the application has contributed and data that has been explicitly shared by another application through appropriate permission granting. If a particular confidential document is mistakenly or accidently indexed using this process, it can lead to a privacy violation by enabling this document to be searched.

Example 1: The following code shows an example of how a private document is indexed on Android using the AppSearch library. This document contains private data that is supposed to be hidden from the indexing process and not allowed to be accessed by AppSearch code.

...
// Document object to index
val doc = Doc(
    namespace="user1",
    id="noteId",
    score=10,
    text="This document contains private data"
)
// Adding document object to AppSearch index
val putRequest = PutDocumentsRequest.Builder().addDocuments(doc).build()