01Description
The application's Cross-Origin Resource Sharing (CORS) policy trusted origins it should not, for example by reflecting any Origin header back in Access-Control-Allow-Origin, trusting the null origin, or matching origins with a loose pattern, while also allowing credentials.
02Impact
A malicious website visited by a logged-in user can make authenticated requests to the application and read the responses, exposing the user's data and any tokens returned by the API.
03Severity
Typically
Medium
Typically Medium. Medium by default. High when the API reflects any origin and allows credentials on endpoints that return sensitive data.
04How to verify
- 1Send requests with arbitrary Origin headers, including attacker-controlled domains, null, and lookalike subdomains.
- 2Check whether the response reflects the origin and includes Access-Control-Allow-Credentials: true.
- 3Confirm sensitive data is returned on the affected endpoints with a benign proof of concept page.
Testing note: only test systems you are authorized to test, within the agreed scope and rules of engagement. Verification steps are written for professional testers and avoid destructive techniques.
05Remediation
- 1Allow only an explicit list of trusted origins, matched exactly.
- 2Never reflect arbitrary origins or trust the null origin when credentials are allowed.
- 3Return CORS headers only on endpoints that need cross-origin access.
Draft findings like this from your own evidence
Book a demo