Mobile app security checklist for founders
Use OWASP MASVS control areas to ask concrete questions about mobile storage, authentication, networking, code and privacy.
The OWASP Mobile Application Security Verification Standard (MASVS) groups mobile security controls into eight areas. We read each area's page on 15 November 2026. Below, OWASP's own control statements are quoted in short form, followed by a question you can ask your team. The questions are our framing, not OWASP's. For the group pages' individual test procedures (the MASTG), we read nothing.
The controls, by area
STORAGE. OWASP says apps handle PII, cryptographic material, secrets and API keys that may be stored locally, in private or in public locations, and may be leaked by accident through misused APIs or system features.
- STORAGE-1: the app securely stores sensitive data. STORAGE-2: the app prevents leakage of sensitive data.
- Ask: what do we keep on the phone, where, and could any of it end up somewhere other apps or the user can read?
CRYPTO. OWASP's reasoning: phones are portable and easily lost, so someone with physical access could reach passwords, financial information and PII. It points to external standards such as NIST SP 800-175B and 800-57 and says poor key management can compromise even strong cryptography.
- CRYPTO-1: current strong cryptography used according to industry best practices. CRYPTO-2: key management according to best practices.
- Ask: which standard libraries do we use, and where are keys generated and stored?
AUTH. OWASP says enforcement must be on the remote endpoint, but the app must follow best practices for the protocols it uses. Some apps rely only on local authentication, which must then be implemented securely. It also says the remote endpoint should be validated against standards such as OWASP ASVS.
- AUTH-1: secure authentication and authorization protocols. AUTH-2: local authentication done securely per platform best practice. AUTH-3: sensitive operations secured with additional authentication.
- Ask: where does the server check what each user may do, and which actions ask for re-authentication?
NETWORK. OWASP says developers typically rely on TLS, but can accidentally disable platform secure defaults or bypass them with low-level APIs or third-party libraries. The area also covers trusting only specific certificate authorities (pinning).
- NETWORK-1: all network traffic secured per current best practices. NETWORK-2: identity pinning for all remote endpoints under the developer's control.
- Ask: is any traffic, including from third-party libraries, outside the secure defaults? Do we pin, and what is the plan when certificates change? (The plan is our question; the page statement is only that the app performs pinning.)
PLATFORM. OWASP covers inter-process communication, WebViews and the user interface. It notes sensitive data such as passwords, card details and one-time passwords can leak through auto-generated screenshots, shoulder surfing or device sharing.
- PLATFORM-1: IPC used securely. PLATFORM-2: WebViews used securely. PLATFORM-3: the user interface used securely.
- Ask: what do we expose to other apps, what do our WebViews load, and what shows on screen or in notifications?
CODE. OWASP says apps receive data from the UI, IPC, network and file system, which should be treated as untrusted, to prevent injection such as SQL injection, XSS or insecure deserialization. It also notes memory corruption flaws are hard to find in penetration testing.
- CODE-1: requires an up-to-date platform version. CODE-2: a mechanism to enforce app updates. CODE-3: only software components without known vulnerabilities. CODE-4: validates and sanitizes all untrusted input.
- Ask: how do we track dependency vulnerabilities, and can we force an old version to update?
RESILIENCE. Defence against reverse engineering and tampering. See our article on mobile app resilience: these are extra layers, not a substitute for server-side security.
- RESILIENCE-1: validates platform integrity. RESILIENCE-2: anti-tampering. RESILIENCE-3: anti-static analysis. RESILIENCE-4: anti-dynamic analysis.
- Ask: does our risk (payments, valuable IP) justify these, and what do they cost in complexity?
PRIVACY. OWASP says MASVS-PRIVACY is a baseline focused on what can be tested from the app itself, using public information or analysis of the app. It is not exhaustive and should not replace assessments required by law, such as a GDPR data protection impact assessment. It gives the example that data collected but not mentioned in the app store listing or privacy policy takes careful manual checking to spot. Data collection and data sharing are treated together, and what happens to data on remote servers is out of MASVS scope. We did not capture this area's individual control statements.
- Ask: what do we collect, does our store listing and privacy policy match what the app really does, and where does it go?
How to use it
- Hold a review. Ask each question and write the answer, even "we do not know".
- Mark each answer confirmed, unknown or gap.
- Fix gaps by risk, credentials and payment data first.
- Repeat before each major release.
Limits
This is a conversation tool, not a security audit or compliance certificate. We read the group pages' overviews and control statements (privacy statements not captured); we did not read individual control pages or the MASTG, and the standard has its own profiles and levels that we do not cover. Apps handling payments, health or children's data may have legal duties beyond this list.
By Omnistu.
Sources
- https://mas.owasp.org/MASVS/05-MASVS-STORAGE/
- https://mas.owasp.org/MASVS/06-MASVS-CRYPTO/
- https://mas.owasp.org/MASVS/07-MASVS-AUTH/
- https://mas.owasp.org/MASVS/08-MASVS-NETWORK/
- https://mas.owasp.org/MASVS/09-MASVS-PLATFORM/
- https://mas.owasp.org/MASVS/10-MASVS-CODE/
- https://mas.owasp.org/MASVS/11-MASVS-RESILIENCE/
- https://mas.owasp.org/MASVS/12-MASVS-PRIVACY/