Software security audit checklist: what to check and why
Short answer
A software security audit is a structured review of an application, its infrastructure and the way it is built and run, measured against a security baseline, to find weaknesses before attackers do. A good checklist covers scope, access control, data, code, dependencies, configuration, logging and response, and ends with a prioritised report and a retest.
An audit differs from a single scan or penetration test. It looks at the whole picture: what exists, who can reach it, how it is built, how it is watched and what happens when something goes wrong. Use the checklist below as a working list. The table that follows maps each area to the NIST Cybersecurity Framework 2.0 so you can show coverage to a board or an auditor.
The checklist, by area
1. Scope and inventory
- List every application, API, service, environment and third-party integration in scope.
- Record data types handled (personal data, payment data, credentials) and where each is stored.
- Name an owner for each asset. An asset with no owner is an audit finding on its own.
2. Threat model and risk
- Draw the trust boundaries: where outside input enters and where data leaves.
- List the likely attackers and what each would want.
- Rank risks by likelihood and impact, so effort goes where it matters.
3. Identity and access control
- Enforce multi-factor authentication on admin and remote access, and prefer phishing-resistant methods where you can.
- Check authorisation on the server for every request, not only in the interface. Broken access control is first on the OWASP Top 10.
- Apply least privilege, review accounts on a schedule and remove leavers the same day.
- Check session handling, password storage and reset flows.
4. Data protection
- Encrypt data in transit with current TLS settings and at rest where it is sensitive.
- Keep secrets out of source code and logs; rotate any that have been exposed.
- Test backups by restoring them, and keep at least one copy that an attacker with production access cannot alter.
- Collect only the personal data you need, and delete it when the purpose ends.
Application, supply chain and operations
5. Application security
- Test against the OWASP Top 10 and, for depth, the OWASP Application Security Verification Standard.
- Check input validation and output encoding for injection and cross-site scripting.
- Check APIs for object-level and function-level authorisation, rate limits and excessive data in responses.
- Run static analysis and dynamic scanning, then a manual review of the findings that tools cannot judge.
6. Dependencies and build pipeline
- Keep an inventory of third-party libraries (a software bill of materials) and scan it for known vulnerabilities.
- Protect the build: restricted access to the repository and pipeline, signed artefacts, reviewed changes. NIST SP 800-218, the Secure Software Development Framework, lists these practices.
- Pin and review updates to dependencies instead of accepting them blindly.
7. Configuration and infrastructure
- Compare servers, containers and cloud accounts against a hardening baseline such as the CIS Benchmarks.
- Close unused ports, disable debug endpoints and default accounts, and check public storage buckets.
- Set security headers and review cross-origin and cookie settings.
- Apply patches within a stated time limit by severity.
8. Logging and monitoring
- Log sign-ins, permission changes and access to sensitive data, and keep the logs where an intruder cannot edit them.
- Alert on events that need a human, and check that the alerts reach someone.
9. Incident response and recovery
- Keep a written response plan with named roles and contact details.
- Rehearse it at least once a year, and record what the rehearsal showed. NIST SP 800-61 describes the incident response lifecycle.
- Define recovery time and recovery point targets, and test them.
10. People and compliance
- Train staff on phishing and on reporting, and test with simulated campaigns.
- Map controls to the frameworks your customers ask about, such as ISO/IEC 27001, SOC 2 or GDPR. Mapping is not certification: say which you hold and which you are working towards.
How the areas map to NIST CSF 2.0
The NIST Cybersecurity Framework 2.0 (2024) organises security outcomes into six functions. Use this table to show which part of the audit covers which function.
| CSF 2.0 function | What it asks | Checklist areas |
|---|---|---|
| Govern | Is security strategy, policy and accountability set? | 1, 2, 10 |
| Identify | Do you know your assets, data and risks? | 1, 2, 6 |
| Protect | Are safeguards in place to limit harm? | 3, 4, 5, 6, 7 |
| Detect | Would you notice an attack or failure? | 8 |
| Respond | Can you act on an incident? | 9 |
| Recover | Can you restore normal service? | 4, 9 |
How to prepare for a software security audit
- Define the objective. Assurance for a customer, a certification such as ISO/IEC 27001, or a general health check each need a different scope.
- Collect documents. Architecture diagrams, asset lists, policies, past audit and test reports, and the release process.
- Name a contact and the stakeholders. One person answers questions and unblocks access.
- Prepare a test environment. Auditors should test a staging copy with realistic data, not production, unless you have agreed otherwise in writing.
- Back up before testing. Active testing can change data.
- Sign the legal terms. NDA, authorisation to test, and a data processing agreement where personal data is involved.
- Agree the timeline and reporting. Milestones, how findings are communicated, and how urgent ones are escalated the same day.
What a useful audit report contains
- Findings with evidence. What was found, where, and the steps to reproduce it.
- Severity and business impact. What an attacker could actually do, in plain language.
- Fix guidance. What to change, ordered by priority.
- Scope and limits. What was not tested, so nobody reads a clean report as a clean bill of health.
- A retest. A second pass confirming the fixes worked. Without it, the audit proves what was wrong, not that it was fixed.
Why independence matters in a security audit
BetterQA is an independent QA company: we test and do not build, so we have no reason to soften a finding about code we wrote. Our security audit combines automated scanning with manual testing, and each finding comes with evidence and fix steps.
Frequently asked questions
Where this comes from
NIST, Cybersecurity Framework 2.0 (2024). NIST SP 800-218, Secure Software Development Framework. NIST SP 800-61, Incident response recommendations. OWASP, Top 10, Web Security Testing Guide and Application Security Verification Standard. CIS Benchmarks. ISO/IEC 27001:2022, Information security management systems.
Want an independent security audit?
Tell us what you run and what a customer or regulator has asked of you. We will scope an audit that fits.
Book a call Security testingAutomate your security testing
Our AI Security Toolkit runs 20 checks we wrote ourselves alongside 15 open-source scanners to find vulnerabilities other scanners miss. SAST, DAST, SCA, secrets detection - unified in one scan.
Explore AI Security Toolkit