I built NoSpamPro because I was tired of "spam blockers" that required uploading my entire contact list or call logs to their servers just to work. Most current solutions are essentially data-harvesting tools disguised as security apps.
NoSpamPro is an Android SMS/Call blocker designed with a privacy-first, local-only philosophy. It uses a hybrid 8-layer filtering system to handle threats without compromising your data.
How it works (The Engineering Side):
95% of the filtering happens on-device. The app follows a decision-tree logic to minimize battery impact and maximize privacy:
Local Heuristics: We use Room DB for instant lookups against known malicious prefixes, B-Codes (Business Sender IDs like B043), and user-defined Regex patterns.
B-Code Registry: A specialized layer that verifies official business codes to detect "spoofing" attempts common in banking fraud.
The AI Layer (Optional): This is where it gets interesting. If the local layers are inconclusive (e.g., a "Likely Spam" score), the app can use Gemini AI for contextual analysis.
Privacy Protocol: Before any text leaves the device, it is anonymized. We only ask the model: "Is this intent-based spam?".
Zero-Log: We implemented a protocol where analysis results are returned and the source data is immediately vaporized. No logs, no archives.
Technical Stack:
- Language: 100% Kotlin. - UI: Jetpack Compose (Material 3). - Storage: Room Persistence Library. - Async: Kotlin Coroutines & Flow.
Why use this instead of the system default? While Google’s default filter is decent, it often misses localized scams (like country-specific betting spam) and has limited customization. NoSpamPro gives you "Advanced Protection" toggles like:
Burst Protection: Detecting "SMS Bomb" attacks and silencing them.
De-shortening URLs: Analyzing the final destination of bit.ly/t.co links within the app's "Ghost Browser."
International Guard: Blocking specific high-risk area codes while whitelisting others.
The app is currently available for Android. I'm looking for feedback specifically on the filtering logic and any edge cases in call screening you might have encountered.
Google Play: https://play.google.com/store/apps/details?id=com.byauth.nos...
I'll be around to answer any technical questions about the implementation or the privacy model!