GHSA-HQ75-XG7R-RX6C
Vulnerability from github – Published: 2025-07-11 17:09 – Updated: 2025-07-11 17:09Summary
Using a CDN that caches (/**/*.png, /**/*.json, /**/*.css, etc...) requests, a cache deception can emerge. This could lead to unauthorized access to user sessions and personal data when cached responses are served to other users.
Details
The vulnerability occurs in the request processing logic where path sanitization is insufficient. The library splits the path using config.basePath but doesn't properly validate the remaining path components. This allows specially crafted requests that appear to be static assets (like /api/auth/get-session/api/auth/image.png assuming config.basePath=/api/auth) to bypass typical CDN cache exclusion rules while actually returning sensitive data.
The problematic code here:
const processRequest = async (request: Request) => {
const url = new URL(request.url);
const path = config?.basePath ? url.pathname.split(config.basePath)[1] : url.pathname;
Since this library is largely coupled with better-auth, it becomes more clear why this can be dangerous with an example request:
Impact
This is a cache deception vulnerability affecting better-call users with CDN caching enabled. which can expose sensitive data.
{
"affected": [
{
"package": {
"ecosystem": "npm",
"name": "better-call"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "1.0.12"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [],
"database_specific": {
"cwe_ids": [
"CWE-525"
],
"github_reviewed": true,
"github_reviewed_at": "2025-07-11T17:09:53Z",
"nvd_published_at": null,
"severity": "MODERATE"
},
"details": "### Summary\n\nUsing a CDN that caches (`/**/*.png`, `/**/*.json`, `/**/*.css`, etc...) requests, a cache deception can emerge. This could lead to unauthorized access to user sessions and personal data when cached responses are served to other users.\n\n### Details\n\nThe vulnerability occurs in the request processing logic where path sanitization is insufficient. The library splits the path using `config.basePath` but doesn\u0027t properly validate the remaining path components. This allows specially crafted requests that appear to be static assets (like `/api/auth/get-session/api/auth/image.png` assuming `config.basePath`=`/api/auth`) to bypass typical CDN cache exclusion rules while actually returning sensitive data.\n\nThe problematic code [here](https://github.com/Bekacru/better-call/blob/8b6f13e24fad7f4666a582601517bb3232d4f4af/src/router.ts#L124):\n```js\n\tconst processRequest = async (request: Request) =\u003e {\n\t\tconst url = new URL(request.url);\n\t\tconst path = config?.basePath ? url.pathname.split(config.basePath)[1] : url.pathname;\n```\n\nSince this library is largely coupled with `better-auth`, it becomes more clear why this can be dangerous with an example request:\n\n\u003cimg width=\"800\" alt=\"image\" src=\"https://github.com/user-attachments/assets/2ab7c4dd-0700-4f59-863f-79f2b5edbb37\" /\u003e\n\n### Impact\n\nThis is a cache deception vulnerability affecting `better-call` users with CDN caching enabled. which can expose sensitive data.",
"id": "GHSA-hq75-xg7r-rx6c",
"modified": "2025-07-11T17:09:53Z",
"published": "2025-07-11T17:09:53Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/Bekacru/better-call/security/advisories/GHSA-hq75-xg7r-rx6c"
},
{
"type": "WEB",
"url": "https://github.com/Bekacru/better-call/commit/7c7d31b"
},
{
"type": "PACKAGE",
"url": "https://github.com/Bekacru/better-call"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N/E:U",
"type": "CVSS_V4"
}
],
"summary": "Better Call routing bug can lead to Cache Deception"
}
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.