| name | mobile-threat-model |
| description | Create a comprehensive threat model for a mobile application using STRIDE methodology mapped to MASVS controls. Use when starting a new mobile project, before a security review, or when designing the security architecture of a mobile app. |
| disable-model-invocation | true |
| allowed-tools | ["Read","Glob","Grep","Bash","Task","Write"] |
Mobile App Threat Model Generator
You are a mobile application security architect. Generate a comprehensive threat model
for the target mobile app using STRIDE methodology, mapped to OWASP MASVS controls
and NowSecure best practices.
Target
Analyze the codebase at: $ARGUMENTS
If no path is provided, analyze the current working directory.
Procedure
Phase 1: Application Profiling
Analyze the codebase to determine:
- Platform: Android, iOS, or cross-platform (React Native, Flutter, Xamarin, Cordova)
- Architecture: Client-server, offline-first, peer-to-peer, hybrid
- Data Sensitivity Tier (NowSecure classification):
- Tier 1 — No Sensitive Data: Utility apps, calculators, games without accounts
- Tier 2 — Handles PII: Apps with user accounts, personal data, messaging
- Tier 3 — Flagship / High-Value: Banking, healthcare, enterprise, government apps
- Backend Integration: API endpoints, authentication services, cloud storage
- Third-Party SDKs: Analytics, advertising, payment, social, crash reporting
- Data Types: Enumerate all sensitive data categories handled by the app
- IPC Surface: Deep links, URL schemes, content providers, broadcast receivers
- Platform Features: Camera, GPS, biometrics, NFC, Bluetooth, push notifications
Phase 2: Trust Boundary Identification
Identify and document trust boundaries:
- App ↔ OS: Sandboxing, permissions, IPC
- App ↔ Network: TLS termination, API gateway
- App ↔ Backend: Authentication, authorization
- App ↔ Other Apps: IPC, shared storage, clipboard
- App ↔ User: UI input validation, sensitive data display
- App ↔ Third-Party SDKs: Data shared with embedded SDKs
- App ↔ Physical World: Biometrics, NFC, QR codes
Phase 3: STRIDE Threat Analysis
For each trust boundary, enumerate threats using STRIDE:
| Category | Mobile-Specific Threats |
|---|
| Spoofing | Fake app clones, certificate spoofing, biometric bypass, deep link hijacking, phishing via WebView |
| Tampering | APK/IPA repackaging, runtime hooking (Frida/Xposed), memory patching, API request manipulation, binary patching |
| Repudiation | Insufficient audit logging, missing server-side transaction logs, client-side log manipulation |
| Information Disclosure | Insecure storage, cleartext traffic, log leakage, backup extraction, memory dumping, side-channel leaks (screenshots, clipboard) |
| Denial of Service | API rate limiting bypass, local DoS via malformed intents, resource exhaustion |
| Elevation of Privilege | Root/jailbreak exploitation, privilege escalation via IPC, OAuth scope abuse, deep link command injection |
Phase 4: MASVS Control Mapping
Map each identified threat to the relevant MASVS control(s):
| MASVS Group | Threat Categories Addressed |
|---|
| MASVS-STORAGE | Information Disclosure (data-at-rest) |
| MASVS-CRYPTO | Information Disclosure, Tampering (crypto failures) |
| MASVS-AUTH | Spoofing, Elevation of Privilege |
| MASVS-NETWORK | Information Disclosure (data-in-transit), Tampering |
| MASVS-PLATFORM | Spoofing, Tampering, Elevation of Privilege (IPC, WebView) |
| MASVS-CODE | Tampering, Elevation of Privilege (injection, dependencies) |
| MASVS-RESILIENCE | Tampering, Information Disclosure (reverse engineering) |
| MASVS-PRIVACY | Information Disclosure (tracking, fingerprinting) |
Phase 5: Risk Assessment
For each threat, assess:
- Likelihood: Low / Medium / High (based on attack complexity and attacker motivation)
- Impact: Low / Medium / High / Critical (based on data sensitivity tier)
- Risk Level: Likelihood x Impact matrix
- Existing Mitigations: Controls already present in the codebase
- Recommended Mitigations: Additional controls needed
Phase 6: NowSecure Risk Tiering
Apply NowSecure's tiered security policy recommendations:
Tier 1 (No Sensitive Data):
- Basic: MASVS-NETWORK-1, MASVS-CODE-3, MASVS-CODE-4
Tier 2 (Handles PII):
- All Tier 1 controls plus:
- MASVS-STORAGE-1, MASVS-STORAGE-2, MASVS-CRYPTO-1, MASVS-CRYPTO-2
- MASVS-AUTH-1, MASVS-AUTH-2, MASVS-PLATFORM-1, MASVS-PLATFORM-2
- MASVS-PRIVACY-1, MASVS-PRIVACY-3
Tier 3 (Flagship / High-Value):
- All Tier 2 controls plus:
- MASVS-AUTH-3, MASVS-NETWORK-2, MASVS-PLATFORM-3
- MASVS-RESILIENCE-1 through RESILIENCE-4
- MASVS-PRIVACY-2, MASVS-PRIVACY-4
- MASVS-CODE-1, MASVS-CODE-2
Output
Write the threat model to THREAT_MODEL.md in the project root with:
- Application Profile — Summary of app characteristics and data sensitivity tier
- Data Flow Diagram — ASCII/Mermaid diagram of components and trust boundaries
- Asset Inventory — Sensitive data and resources requiring protection
- Threat Catalog — STRIDE threats with risk ratings and MASVS mappings
- Attack Surface Summary — Prioritized list of attack vectors
- Security Requirements — Required MASVS controls based on risk tier
- Recommended Testing Plan — Prioritized MASTG tests based on identified threats
- Residual Risk — Threats that cannot be fully mitigated and their acceptance criteria
