Security Misconfiguration
Key Takeaways
- Security misconfiguration is an insecure setting, default or omission anywhere in the stack, not a flaw in application code.
- It ranks A02 in the OWASP Top 10:2025, up from A05 in 2021.
- OWASP found some form of misconfiguration in 100% of the applications tested for the 2025 edition.
- Repeatable security hardening and continuous verification beat one-time audits.
What Is Security Misconfiguration?
Security misconfiguration is a weakness created when a system, application or cloud service is deployed in a less secure state than it should be. The code can be fine. The risk comes from a setting left at its default, a feature left on or a protection never enabled, at any layer from the network to the framework to the cloud account.
The Most Common Security Misconfiguration Types
- Default credentials and sample content: vendor passwords, sample apps and test pages left in production.
- Unnecessary exposure: open ports, admin consoles, debug endpoints and directory listing.
- Verbose errors and missing headers: stack traces that reveal versions, and no Content-Security-Policy or Strict-Transport-Security headers.
- Cloud misconfiguration: public storage buckets, overly broad IAM roles and open security groups.
- Pipeline settings: Novee’s Cordyceps research showed how GitHub Actions workflows that mishandle untrusted input can hand outsiders control of repositories and cloud credentials.
Why Security Misconfiguration Ranks #2 on the OWASP Top 10
Security misconfiguration is A02 in the OWASP Top 10:2025, up from A05 in 2021, A6 in 2017 and A5 in 2013. It has never held the top spot, which belongs to Broken Access Control in 2021 and 2025, but its reach is unmatched: OWASP reports misconfiguration in 100% of tested applications and over 719,000 mapped CWE occurrences.
OWASP’s explanation is that more application behavior is now driven by configuration rather than code. Frameworks, cloud services and containers expose hundreds of settings, and each is a chance to choose the insecure option.
Why Misconfiguration Keeps Appearing Despite Being Preventable
Defaults favor convenience
Products often ship open so they work on first install, leaving hardening to busy customers.
Environments drift
Hotfixes and manual console changes pull production away from the reviewed baseline.
Ownership is unclear
Under the cloud shared responsibility model, app and platform teams each assume the other secured the setting.
How to Detect and Remediate Security Misconfiguration
Build a hardened baseline
Make security hardening a repeatable, automated process so every environment is configured identically, as OWASP recommends. Remove unused components and replace all default credentials before deployment.
Scan continuously
Configuration scanners catch known-bad settings, and DAST finds runtime issues like missing headers and exposed admin paths.
Validate what is exploitable
Novee’s continuous AI pentesting covers web apps, APIs and the external attack surface, reasoning about how an exposed setting chains with permissions and business logic. Three independent agents validate each finding, and Agentic Remediation retests every fix.
FAQ
Not exactly. A software vulnerability is usually a flaw in code that exists wherever that version runs. A security misconfiguration lives in how a specific deployment is set up, so the same software can be safe in one environment and exposed in another. Both are exploitable weaknesses, but they are fixed differently.
Partly. DAST reliably flags externally visible issues such as missing security headers, verbose error pages, directory listing and some exposed admin interfaces. It cannot see internal settings like IAM policies, database permissions or storage access rules, and it rarely shows whether a finding is exploitable. Pair it with configuration scanning and manual or AI-driven testing.
It affects APIs just as much. Common API misconfigurations include permissive CORS policies, debug endpoints left enabled, missing rate limits, exposed API documentation in production and gateways that skip authentication on certain routes. Because APIs often connect straight to data stores, one misconfigured endpoint can expose more data than a web page would.
Infrastructure-as-code turns settings into reviewable, versioned files. Teams can scan templates for insecure values before deployment, apply the same hardened baseline to every environment, and roll back bad changes quickly. It does not remove risk on its own, since a flawed template spreads the same mistake everywhere, but it makes errors visible and repeatable fixes possible.
Drift happens when live systems change outside the approved process, through manual console edits, emergency fixes or temporary debug settings that never get reverted. Each change moves production further from the reviewed baseline. Without continuous comparison against the intended configuration, these gaps accumulate quietly until an attacker or audit finds them.