Roadmap

Future Roadmap

HydraDragonAV Mobile is an evolving project with several planned improvements.

Detection Improvements

• Enhanced ML Models: Training on larger and more diverse APK datasets, including more obfuscation techniques and adware variants.
• Additional YARA Rules: Community-contributed rules targeting emerging Android malware families.
• Improved Unicorn Emulation: Broader syscall emulation for more complete native code sandboxing.

Performance & Stability

• Reduced Memory Footprint: Further optimization of the XOR filter memory usage.
• Faster Database Loading: Streaming and lazy-loading improvements for signature databases.
• Battery Optimization: More efficient screen capture scheduling and reduced wake frequency.

Feature Plans

Feature	Status
Scheduled scans (daily/weekly at specific times)	Planned
Custom scan exclusions (per-folder, per-app)	Planned
Export scan reports (JSON/CSV)	Planned
More granular per-detector thresholds in Settings	Planned
Additional languages for on-screen text scanning	Under consideration
Dark mode / Material You theming	Planned
App lock / PIN protection for Settings	Planned

Community & Open Source

• Better Documentation: Continuous updates to this wiki and inline code documentation.
• YARA Rule Repository: Separate repository for community-contributed YARA rules with PR-based review.
• CI/CD Improvements: More comprehensive automated testing across Android versions and device types.

Long-Term Research

• On-device LLM-based analysis: Exploring small language models for understanding app behavior descriptions.
• Graph-based behavior analysis: Mapping app behavior as execution graphs for anomaly detection.
• Side-channel detection: Using CPU/GPU temperature and power draw anomalies to detect hidden mining malware.

See Also

• Known-Limitations — Current limitations
• Contributing — How to help with these goals