CORS
CORS is a controlled relaxation of the same-origin policy
Last updated
CORS is a controlled relaxation of the same-origin policy
Last updated
MANUALY TESTING FOR CORS:
ARBITRARY ORIGINS: Change the origin to an arbitrary vaule / true
GET: Origin: https://malicious-website.com
POST: Access-Control-Allow-Origin: https://malicious-website.com
MULL ORIGINS: Change the origin header to the null value
GET: Origin: null
POST: Access-Control-Allow-Origin: null
SAME ORIGINS: Change the origin header to one that begins with the origin of the site
GET: Origin: https://website.com
POST: Access-Control-Allow-Origin: https://website.com
SUBDOMAIN ORIGINS: Change the origin to one that ends with the origin of the site.
GET: Origin: https://random.website.com
POST: Access-Control-Allow-Origin: https://random.website.com
Same-origin policy
The same-origin policy is a restrictive cross-origin specification that limits the ability for a website to interact with resources outside of the source domain. The same-origin policy restricts scripts on one origin from accessing data from another origin. An origin consists of a URI scheme, domain and port number. For example, consider the following URL:
How is the same-origin policy implemented?
The same-origin policy generally controls the access that JavaScript code has to content that is loaded cross-domain. Cross-origin loading of page resources is generally permitted. For example, the SOP allows embedding of images via the < img > tag, media via the tag and JavaScript includes with the tag. However, while these external resources can be loaded by the page, any JavaScript on the page won't be able to read the contents of these resources.
There are various exceptions to the same-origin policy:
Some objects are writable but not readable cross-domain, such as the location object or the location.href property from iframes or new windows.
Some objects are readable but not writable cross-domain, such as the length property of the window object (which stores the number of frames being used on the page) and the closed property.
The replace function can generally be called cross-domain on the location object.
You can call certain functions cross-domain. For example, you can call the functions close, blur and focus on a new window. The postMessage function can also be called on iframes and new windows in order to send messages from one domain to another.
Due to legacy requirements, the same-origin policy is more relaxed when dealing with cookies, so they are often accessible from all subdomains of a site even though each subdomain is technically a different origin. You can partially mitigate this risk using the HttpOnly cookie flag. It's possible to relax same-origin policy using document.domain. This special property allows you to relax SOP for a specific domain, but only if it's part of your FQDN (fully qualified domain name). For example, you might have a domain marketing.example.com and you would like to read the contents of that domain on example.com. To do so, both domains need to set document.domain to example.com. Then SOP will allow access between the two domains despite their different origins. In the past it was possible to set document.domain to a TLD such as com, which allowed access between any domains on the same TLD, but now modern browsers prevent this.
Relaxation of the same-origin policy
The same-origin policy is very restrictive and consequently various approaches have been devised to circumvent the constraints. Many websites interact with subdomains or third-party sites in a way that requires full cross-origin access. A controlled relaxation of the same-origin policy is possible using cross-origin resource sharing (CORS).
The cross-origin resource sharing protocol uses a suite of HTTP headers that define trusted web origins and associated properties such as whether authenticated access is permitted. These are combined in a header exchange between a browser and the cross-origin web site that it is trying to access.
CORS and the Access-Control-Allow-Origin response header
Access-Control-Allow-Origin The cross-origin resource sharing specification provides controlled relaxation of the same-origin policy for HTTP requests to one website domain from another through the use of a collection of HTTP headers. Browsers permit access to responses to cross-origin requests based upon these header instructions.
The Access-Control-Allow-Origin header is included in the response from one website to a request originating from another website, and identifies the permitted origin of the request. A web browser compares the Access-Control-Allow-Origin with the requesting website's origin and permits access to the response if they match.
Implementing simple CORS
Access-Control-Allow-Origin
The cross-origin resource sharing (CORS) specification prescribes header content exchanged between web servers and browsers that restricts origins for web resource requests outside of the origin domain. The CORS specification identifies a collection of protocol headers of which Access-Control-Allow-Origin is the most significant. This header is returned by a server when a website requests a cross-domain resource, with an Origin header added by the browser.
For example, suppose a website with origin normal-website.com causes the following cross-domain request:
The server on robust-website.com returns the following response:
The browser will allow code running on normal-website.com to access the response because the origins match. The specification of Access-Control-Allow-Origin allows for multiple origins, or the value null, or the wildcard *. However, no browser supports multiple origins and there are restrictions on the use of the wildcard *.
Handling cross-origin resource requests with credentials
Access-Control-Allow-Credentials: true
The default behavior of cross-origin resource requests is for requests to be passed without credentials like cookies and the Authorization header. However, the cross-domain server can permit reading of the response when credentials are passed to it by setting the CORS Access-Control-Allow-Credentials header to true. Now if the requesting website uses JavaScript to declare that it is sending cookies with the request:
And the response to the request is:
Then the browser will permit the requesting website to read the response, because the Access-Control-Allow-Credentials response header is set to true. Otherwise, the browser will not allow access to the response.
Relaxation of CORS specifications with wildcards
Access-Control-Allow-Origin: supports wildcards Access-Control-Allow-Origin: *
Fortunately, from a security perspective, the use of the wildcard is restricted in the specification as you cannot combine the wildcard with the cross-origin transfer of credentials (authentication, cookies or client-side certificates). Consequently, a cross-domain server response of the form:
is not permitted as this would be dangerously insecure, exposing any authenticated content on the target site to everyone. Given these constraints, some web servers dynamically create Access-Control-Allow-Origin headers based upon the client-specified origin. This is a workaround for CORS constraints that is not secure.
Pre-flight checks
Special-Request-Header
//CORS configuration issues
Many modern websites use CORS to allow access from subdomains and trusted third parties. Their implementation of CORS may contain mistakes or be overly lenient to ensure that everything works, and this can result in exploitable vulnerabilities.
Server-generated Access-Control-Allow-Origin(ACAO) header from client-specified Origin header Some applications need to provide access to a number of other domains. Maintaining a list of allowed domains requires ongoing effort, and any mistakes risk breaking functionality. So some applications take the easy route of effectively allowing access from any other domain.
One way to do this is by reading the Origin header from requests and including a response header stating that the requesting origin is allowed. For example, consider an application that receives the following request:
It then responds with:
These headers state that access is allowed from the requesting domain (malicious-website.com) and that the cross-origin requests can include cookies (Access-Control-Allow-Credentials: true) and so will be processed in-session. Because the application reflects arbitrary origins in the Access-Control-Allow-Origin header, this means that absolutely any domain can access resources from the vulnerable domain. If the response contains any sensitive information such as an API key or CSRF token, you could retrieve this by placing the following script on your website:
//LAB: CORS vulnerability with basic origin reflection
Access-Control-Allow-Origin: true
craft some JavaScript that uses CORS to retrieve the administrator's API key and upload the code to your exploit server.
MANUALY TESTING FOR CORS:
1. ARBITRARY ORIGINS: Change the origin to an arbitrary vaule / true
GET: Origin: https://malicious-website.com
POST: Access-Control-Allow-Origin: https://malicious-website.com
2. Change the origin header to the null value
GET: Origin: null
POST: Access-Control-Allow-Origin: null
Errors parsing Origin headers
Some applications that support access from multiple origins do so by using a whitelist of allowed origins. When a CORS request is received, the supplied origin is compared to the whitelist. If the origin appears on the whitelist then it is reflected in the Access-Control-Allow-Origin header so that access is granted. The application checks the supplied origin against its list of allowed origins and, if it is on the list
Mistakes often arise when implementing CORS origin whitelists. Some organizations decide to allow access from all their subdomains (including future subdomains not yet in existence). And some applications allow access from various other organizations' domains including their subdomains. These rules are often implemented by matching URL prefixes or suffixes, or using regular expressions.
Mistakes often arise when implementing CORS origin whitelists. Some organizations decide to allow access from all their subdomains (including future subdomains not yet in existence). And some applications allow access from various other organizations' domains including their subdomains. These rules are often implemented by matching URL prefixes or suffixes, or using regular expressions.
Any mistakes in the implementation can lead to access being granted to unintended external domains. For example, suppose an application grants access to all domains ending in:
An attacker might be able to gain access by registering the domain:
Alternatively, suppose an application grants access to all domains beginning with:
An attacker might be able to gain access using the domain:
Whitelisted null origin value
The specification for the Origin header supports the value null. Browsers might send the value null in the Origin header in various unusual situations:
Cross-origin redirects.
Requests from serialized data.
Request using the file: protocol.
Sandboxed cross-origin requests.
Some applications might whitelist the null origin to support local development of the application. For example, suppose an application receives the following cross-origin request:
In this situation, an attacker can use various tricks to generate a cross-origin request containing the value null in the Origin header. This will satisfy the whitelist, leading to cross-domain access. For example, this can be done using a sandboxed iframe cross-origin request of the form:
//Lab: CORS vulnerability with trusted null origin
Exploiting XSS via CORS trust relationships
Even "correctly" configured CORS establishes a trust relationship between two origins. If a website trusts an origin that is vulnerable to cross-site scripting (XSS), then an attacker could exploit the XSS to inject some JavaScript that uses CORS to retrieve sensitive information from the site that trusts the vulnerable application.
Given the following request:
If the server responds with:
Then an attacker who finds an XSS vulnerability on subdomain.vulnerable-website.com could use that to retrieve the API key, using a URL like:
Breaking TLS with poorly configured CORS
Suppose an application that rigorously employs HTTPS also whitelists a trusted subdomain that is using plain HTTP. For example, when the application receives the following request:
The application responds with:
In this situation, an attacker who is in a position to intercept a victim user's traffic can exploit the CORS configuration to compromise the victim's interaction with the application. This attack involves the following steps:
The victim user makes any plain HTTP request.
The attacker injects a redirection to: http://trusted-subdomain.vulnerable-website.com
The victim's browser follows the redirect.
The attacker intercepts the plain HTTP request, and returns a spoofed response containing a CORS request to: https://vulnerable-website.com
The victim's browser makes the CORS request, including the origin: http://trusted-subdomain.vulnerable-website.com
The application allows the request because this is a whitelisted origin. The requested sensitive data is returned in the response.
The attacker's spoofed page can read the sensitive data and transmit it to any domain under the attacker's control. This attack is effective even if the vulnerable website is otherwise robust in its usage of HTTPS, with no HTTP endpoint and all cookies flagged as secure.
//Lab: CORS vulnerability with trusted insecure protocols
Intranets and CORS without credentials
Most CORS attacks rely on the presence of the response header: Access-Control-Allow-Credentials: true Without that header, the victim user's browser will refuse to send their cookies, meaning the attacker will only gain access to unauthenticated content, which they could just as easily access by browsing directly to the target website.
However, there is one common situation where an attacker can't access a website directly: when it's part of an organization's intranet, and located within private IP address space. Internal websites are often held to a lower security standard than external sites, enabling attackers to find vulnerabilities and gain further access. For example, a cross-origin request within a private network may be as follows:
And the server responds with:
The application server is trusting resource requests from any origin without credentials. If users within the private IP address space access the public internet then a CORS-based attack can be performed from the external site that uses the victim's browser as a proxy for accessing intranet resources.
//Lab: CORS vulnerability with internal network pivot attack
How to prevent CORS-based attacks
CORS vulnerabilities arise primarily as misconfigurations. Prevention is therefore a configuration problem. The following sections describe some effective defenses against CORS attacks.
Proper configuration of cross-origin requests: If a web resource contains sensitive information, the origin should be properly specified in the Access-Control-Allow-Origin header.
Only allow trusted sites: It may seem obvious but origins specified in the Access-Control-Allow-Origin header should only be sites that are trusted. In particular, dynamically reflecting origins from cross-origin requests without validation is readily exploitable and should be avoided.
Avoid whitelisting null: Avoid using the header Access-Control-Allow-Origin: null. Cross-origin resource calls from internal documents and sandboxed requests can specify the null origin. CORS headers should be properly defined in respect of trusted origins for private and public servers.
Avoid wildcards in internal networks: Avoid using wildcards in internal networks. Trusting network configuration alone to protect internal resources is not sufficient when internal browsers can access untrusted external domains.
CORS is not a substitute for server-side security policies: CORS defines browser behaviors and is never a replacement for server-side protection of sensitive data - an attacker can directly forge a request from any trusted origin. Therefore, web servers should continue to apply protections over sensitive data, such as authentication and session management, in addition to properly configured CORS
