A web application firewall (WAF) inspects HTTP and HTTPS traffic to a web application and blocks requests matching known attack patterns, such as SQL injection and cross site scripting. It operates at the application layer, unlike a network firewall that filters by port and protocol. A WAF is a compensating control, reducing exposure to common attacks, but it does not fix underlying application vulnerabilities.
A network firewall decides whether a connection to a web server is allowed at all. It has nothing to say about whether the request inside that connection is trying to steal a database through a malformed form field. That gap is what a web application firewall exists to close.
What does a web application firewall inspect?
A web application firewall (WAF) stands in the path between users and a web application, typically as a reverse proxy, a cloud service, or a module within a load balancer or content delivery network. It reads the actual content of each HTTP or HTTPS request, including URLs, headers, cookies and form data, and compares that content against rules designed to spot malicious patterns.
Where a network firewall asks whether a connection to port 443 from a given address should be allowed, a WAF asks whether the data inside that allowed connection looks like an attack. A request that includes a string resembling an SQL command where a product ID was expected, or a script tag where a comment was expected, gets flagged or blocked before it reaches the application.
Most WAFs operate on a rule based model, often built on widely used rule sets that codify known attack signatures, supplemented by custom rules tuned to the specific application. Some also incorporate anomaly detection or machine learning models that learn a baseline of normal traffic and flag deviations, though rule based detection of known patterns remains the core mechanism.
What attacks does a WAF stop?
The OWASP Top 10, a widely referenced list of common web application risks, maps closely to what WAFs are built to catch. Injection attacks, where an attacker inserts malicious code into an input field that the application then executes, are a primary target, covering SQL injection against a database and command injection against an operating system.
Cross site scripting (XSS), where an attacker injects a script that runs in another user’s browser, is another common target, along with cross site request forgery, which tricks an authenticated user’s browser into submitting a request the user did not intend. WAFs also commonly block attempts to exploit known vulnerabilities in popular content management systems, plugins and frameworks, often faster than every affected site can patch.
Beyond specific attack patterns, many WAFs provide rate limiting, throttling requests from a single source to slow brute force login attempts and credential stuffing, and bot management, distinguishing legitimate users from automated scraping, scalping and abuse tools.
What can’t a WAF do?
A WAF is a compensating control, not a fix. It reduces the likelihood that a known attack pattern reaches a vulnerable piece of code, but the vulnerability itself remains. If the WAF is misconfigured, bypassed through an encoding trick, or simply does not have a rule for a novel technique, the underlying flaw is exposed exactly as before.
Business logic flaws are a particular blind spot. A WAF looks for malicious patterns in individual requests, but it generally cannot detect a sequence of entirely valid looking requests that together exploit a flaw in the application’s logic, such as manipulating a shopping cart to apply a discount twice, or accessing another user’s data by changing an ID in an otherwise well formed request. These require the kind of manual and automated testing performed during penetration testing.
Overly aggressive WAF rules also create their own problem: false positives that block legitimate traffic, frustrating users and sometimes prompting teams to loosen rules to the point where they stop being useful. Tuning a WAF against real traffic patterns, rather than deploying default rules and leaving them unreviewed, is ongoing work, not a one off setup task.
How does a WAF fit with other controls?
A WAF is one layer in a broader approach to protecting a web application, working alongside secure coding practices, regular vulnerability management, and periodic penetration testing that specifically probes for the logic flaws and novel techniques a WAF cannot catch. This layering reflects the principle of defence in depth: no single control, including a WAF, is expected to be the only thing standing between an attacker and the application.
For organisations that also run intrusion detection or a SIEM, WAF logs are a valuable feed, showing what attack attempts are being made against public facing applications even when they are successfully blocked, which helps prioritise where deeper testing and remediation effort should go.
Where does a WAF matter for Australian organisations?
Australian organisations running customer facing applications, particularly those handling payment card data or personal information, commonly deploy a WAF as part of their broader control set. Under PCI DSS, a WAF or an equivalent process of regular application security testing is an explicit option for protecting public facing web applications that process card data.
For organisations pursuing an IRAP assessment or building towards government security requirements, WAF presence and configuration is one of the areas commonly examined when assessing exposure of internet facing systems, though it is assessed alongside the underlying application’s own security posture rather than treated as a substitute for it.
Cybernion’s penetration testing services test web applications with and without WAF protection in scope, to establish both what an external attacker who has to get past the WAF would face, and what the application’s own security looks like underneath it.
Frequently asked questions
What is a web application firewall (WAF)?
A web application firewall (WAF) is a security control that stands in front of a web application and inspects HTTP and HTTPS requests and responses, looking for patterns associated with common web attacks. It blocks or flags requests that match known attack signatures, such as SQL injection attempts or cross site scripting payloads, before they reach the application. A WAF operates at the application layer, unlike a network firewall, which filters based on ports and protocols.
What is the difference between a WAF and a regular firewall?
A regular network firewall filters traffic based on IP address, port and protocol, deciding which connections are allowed to reach a server at all. A web application firewall inspects the actual content of HTTP requests reaching a web application, looking for malicious patterns within otherwise permitted traffic, such as an SQL injection string hidden inside a form field. The two operate at different layers and are complementary rather than substitutes for each other.
What attacks does a WAF protect against?
A WAF primarily targets attacks described in resources like the OWASP Top 10, including SQL injection, cross site scripting, cross site request forgery, and attempts to exploit known vulnerabilities in common web frameworks and plugins. Many WAFs also provide rate limiting to slow credential stuffing and brute force attempts, and bot management features to filter automated scraping and abuse traffic separately from genuine users.
Does a WAF replace the need for secure coding?
No. A WAF is a compensating control that catches known attack patterns at the network edge, but it cannot fix an underlying vulnerability in the application’s code, such as improper input validation or a flawed authorisation check. Relying on a WAF alone leaves an organisation exposed to novel attack patterns the WAF does not recognise, and to logic flaws a WAF cannot detect at all. Secure development practices and regular testing remain necessary.
How does a WAF fit into an Australian organisation’s security stack?
Australian organisations running customer facing web applications, particularly those handling payment data or personal information, commonly deploy a WAF as part of a defence in depth approach. It supports PCI DSS requirements for organisations handling card data, and is one control an IRAP assessor or penetration tester will note the presence and configuration of when reviewing a system’s exposure to external attack, alongside the underlying application’s own security controls.
Sources:
- OWASP Top 10, OWASP Foundation, 2025
- PCI DSS Requirements and Testing Procedures, PCI Security Standards Council, 2024
- Guidelines for Software Development, Information Security Manual, Australian Signals Directorate, 2026
Last updated: 30 August, 2026
