Zimperium Reveals New Android Malware Evasion Techniques and Risks

Alan V Gutnov
Alan V Gutnov

Director of Strategy

 
July 16, 2025
3 min read

Konfety Malware Overview

The new Konfety malware variant targets Android devices using advanced evasion techniques, complicating detection and analysis for security professionals. This malware employs an "evil-twin" tactic to conduct fraudulent activities by mimicking legitimate applications.

Key features of this variant include:

  • Dual-App Deception: Both the benign and malicious versions share the same package name, enhancing the malware's evasiveness.
  • ZIP-Level Evasion: The APK structure is tampered with, including unsupported compression and fake encryption flags, complicating reverse engineering.
  • Dynamic Code Loading: Key functionalities are concealed in encrypted assets, only decrypted at runtime.
  • Stealth Techniques: This includes hiding the app icon, mimicking legitimate apps, and using geofencing to adjust behavior based on the user's location.
  • Ad Fraud Infrastructure: Utilizes the CaramelAds SDK for ad fetching and maintaining communication with attacker-controlled servers.

The sophistication of the Konfety malware is evident in its constant adaptation to evade detection.

Evasion Techniques

Image courtesy of Zimperium

Evasion via Malformed ZIP Packaging

Recent variants of Konfety employ advanced techniques to avoid detection, including:

  • General Purpose Flag Enabled: This causes analysis tools to mistakenly identify the APK as encrypted, requiring a password for decompression.
  • Unsupported Compression Method (BZIP - 0x000C): The AndroidManifest.xml falsely declares the use of BZIP compression, leading to analysis tool failures.

These manipulations can prevent tools like APKTool and JADX from executing properly, further complicating the detection of the malware.

APKTool Failure

Image courtesy of Zimperium

Analysis Failure

Image courtesy of Zimperium

The Konfety Malware Analysis

Konfety incorporates multiple layers of obfuscation to hinder detection efforts:

  • Dynamic Code Loading: It loads additional executable code at runtime from encrypted assets, making it difficult to detect critical functionalities during scans.

Dynamic Code

Image courtesy of Zimperium

  • Decoy Applications: The malware mimics legitimate apps with the same package name, emphasizing its stealthy intent.

Decoy Application

Image courtesy of Zimperium

Network Traffic Analysis

Through dynamic analysis, it was observed that the malware establishes a browser connection to malicious websites after the user accepts a User Agreement. This leads to a series of redirects designed to trick users into installing additional malicious applications or subscribing to unwanted notifications.

Network Traffic

Image courtesy of Zimperium

Zimperium Protection Against Konfety

Zimperium's on-device Mobile Threat Defense (MTD) solution provides comprehensive protection against Konfety malware. The platform continuously adapts to evolving threat landscapes, ensuring that users remain secure from sophisticated malware such as Konfety.

For further information on how Zimperium can safeguard your mobile environment, visit Zimperium.

Unsupported Compression Methods

Zimperium researchers identified that 3,300 Android malware samples use unsupported compression methods to bypass detection. This technique limits analysis capabilities and is particularly effective against tools like JADX and APKTool.

Unsupported Compression

Image courtesy of Zimperium

Zimperium's Detection Capabilities

Zimperium maintains dynamic on-device threat detection engines that protect against malicious apps installed using unsupported compression methods.

Zimperium MTD's enhanced phishing detection can also prevent phishing attacks, offering robust protection against the full attack chain.

Explore how our solutions can secure your devices and protect your business from evolving mobile threats. For more information, visit Zimperium.

Alan V Gutnov
Alan V Gutnov

Director of Strategy

 

MBA-credentialed cybersecurity expert specializing in Post-Quantum Cybersecurity solutions with proven capability to reduce attack surfaces by 90%.

Related News

New U.S. Directive Sets 2030 Deadline for Contractor Transition to Quantum-Resistant Cryptography
post-quantum cryptography standards

New U.S. Directive Sets 2030 Deadline for Contractor Transition to Quantum-Resistant Cryptography

New federal mandate requires contractors to adopt NIST-approved quantum-resistant cryptography by 2030 to combat 'harvest now, decrypt later' cyber threats.

By Alan V Gutnov July 27, 2026 4 min read
common.read_full_article
New Industry Report Outlines Strategic Defense Frameworks Against Harvest Now, Decrypt Later Quantum Threats
harvest now decrypt later

New Industry Report Outlines Strategic Defense Frameworks Against Harvest Now, Decrypt Later Quantum Threats

Learn how 'Harvest Now, Decrypt Later' quantum threats endanger your data. Discover essential NIST post-quantum migration strategies for your enterprise.

By Brandon Woo July 24, 2026 4 min read
common.read_full_article
NIST Finalizes 2026 Technical Requirements for Post-Quantum Cryptographic Infrastructure Migration and Implementation
NIST post-quantum cryptography standards

NIST Finalizes 2026 Technical Requirements for Post-Quantum Cryptographic Infrastructure Migration and Implementation

NIST finalizes 2026 technical requirements for post-quantum cryptographic migration. Learn the key FIPS standards essential for quantum-resistant infrastructure.

By Alan V Gutnov July 23, 2026 5 min read
common.read_full_article
NIST Post-Quantum Cryptography Standards Formally Adopted for Model Context Protocol Infrastructure Deployments
NIST post-quantum cryptography standards

NIST Post-Quantum Cryptography Standards Formally Adopted for Model Context Protocol Infrastructure Deployments

NIST has finalized post-quantum cryptographic standards. Learn how to implement ML-KEM and ML-DSA to secure your Model Context Protocol (MCP) infrastructure.

By Brandon Woo July 22, 2026 5 min read
common.read_full_article