Axios has Unrestricted Cloud Metadata Exfiltration via Header Injection Chain
Vulnerability Disclosure: Unrestricted Cloud Metadata Exfiltration via Header Injection Chain
Summary
The Axios library is vulnerable to a specific gadget-style attack chain in which prototype pollution in a third-party dependency may be leveraged to inject unsanitized header values into outbound requests.
Axios can be used as a gadget after pollution occurs elsewhere because header values merged from attacker-controlled prototype properties are not sanitized for CRLF (\r\n) characters before being written to the request. In affected deployments, this may enable limited request manipulation or metadata access as part of a higher-complexity exploit chain.
Severity: Moderate (CVSS 3.1 Base Score: 4.8)
Affected Versions: All versions (v0.x - v1.x)
Vulnerable Component: lib/adapters/http.js (Header Processing)
Usage of \"Helper\" Vulnerabilities
This issue requires a separate prototype pollution vulnerability in another library in the application stack (for example, qs, minimist, ini, or body-parser). If an attacker can pollute Object.prototype, Axios may pick up the polluted properties during config merge.
Because Axios does not sanitise these merged header values for CRLF (\r\n) characters, the polluted property can alter the structure of an outbound HTTP request.
Proof of Concept
1. The Setup (Simulated Pollution)
Imagine a scenario where a known vulnerability exists in a query parser. The attacker sends a payload that sets:
``javascript
Object.prototype['x-amz-target'] = \"dummy\r\n\r\nPUT /latest/api/token HTTP/1.1\r\nHost: 169.254.169.254\r\nX-aws-ec2-metadata-token-ttl-seconds: 21600\r\n\r\nGET /ignore\";
`
2. The Gadget Trigger (Safe Code)
The application makes a completely safe, hardcoded request:
`javascript
// This looks safe to the developer
await axios.get('https://analytics.internal/pings');
`
3. The Execution
Axios merges the prototype property x-amz-target into the request headers. It then writes the header value directly to the socket without validation.
Resulting HTTP traffic:
`http
GET /pings HTTP/1.1
Host: analytics.internal
x-amz-target: dummy
PUT /latest/api/token HTTP/1.1
Host: 169.254.169.254
X-aws-ec2-metadata-token-ttl-seconds: 21600
GET /ignore HTTP/1.1
...
`
4. The Impact
In environments where requests can reach cloud metadata endpoints or sensitive internal services, the injected header content may help bypass expected request constraints and expose limited credentials or modify request semantics. This impact depends on application context and a separate prototype-pollution primitive.
Impact Analysis
Confidentiality: May expose limited sensitive information in affected network environments.
Integrity: May allow modification of outbound request structure or injected headers.
Attack Complexity: Exploitation requires a separate prototype-pollution vulnerability and a reachable target service.
Recommended Fix
Validate all header values in lib/adapters/http.js and xhr.js before passing them to the underlying request function.
Patch Suggestion:
`javascript
// In lib/adapters/http.js
utils.forEach(requestHeaders, function setRequestHeader(val, key) {
if (/[\r\n]/.test(val)) {
throw new Error('Security: Header value contains invalid characters');
}
// ... proceed to set header
});
``
References
OWASP: CRLF Injection (CWE-113)
This report was generated as part of a security audit of the Axios library.
lodash vulnerable to Prototype Pollution via array path bypass in `_.unset` and `_.omit`
Impact
Lodash versions 4.17.23 and earlier are vulnerable to prototype pollution in the _.unset and _.omit functions. The fix for CVE-2025-13465 only guards against string key members, so an attacker can bypass the check by passing array-wrapped path segments. This allows deletion of properties from built-in prototypes such as Object.prototype, Number.prototype, and String.prototype.
The issue permits deletion of prototype properties but does not allow overwriting their original behavior.
Patches
This issue is patched in 4.18.0.
Workarounds
None. Upgrade to the patched version.
Next.js: HTTP request smuggling in rewrites
Summary
When Next.js rewrites proxy traffic to an external backend, a crafted DELETE/OPTIONS request using Transfer-Encoding: chunked could trigger request boundary disagreement between the proxy and backend. This could allow request smuggling through rewritten routes.
Impact
An attacker could smuggle a second request to unintended backend routes (for example, internal/admin endpoints), bypassing assumptions that only the configured rewrite destination/path is reachable. This does not impact applications hosted on providers that handle rewrites at the CDN level, such as Vercel.
Patches
The vulnerability originated in an upstream library vendored by Next.js. It is fixed by updating that dependency’s behavior so content-length: 0 is added only when both content-length and transfer-encoding are absent, and transfer-encoding is no longer removed in that code path.
Workarounds
If upgrade is not immediately possible:
Block chunked DELETE/OPTIONS requests on rewritten routes at your edge/proxy.
Enforce authentication/authorization on backend routes per our security guidance.
Next.js: Unbounded next/image disk cache growth can exhaust storage
Summary
The default Next.js image optimization disk cache (/_next/image) did not have a configurable upper bound, allowing unbounded cache growth.
Impact
An attacker could generate many unique image-optimization variants and exhaust disk space, causing denial of service. Note that this does not impact platforms that have their own image optimization capabilities, such as Vercel.
Patches
Fixed by adding an LRU-backed disk cache with images.maximumDiskCacheSize, including eviction of least-recently-used entries when the limit is exceeded. Setting maximumDiskCacheSize: 0 disables disk caching.
Workarounds
If upgrade is not immediately possible:
Periodically clean .next/cache/images.
Reduce variant cardinality (e.g., tighten values for images.localPatterns, images.remotePatterns, and images.qualities)
Axios is Vulnerable to Denial of Service via __proto__ Key in mergeConfig
Denial of Service via proto Key in mergeConfig
Summary
The mergeConfig function in axios crashes with a TypeError when processing configuration objects containing __proto__ as an own property. An attacker can trigger this by providing a malicious configuration object created via JSON.parse(), causing complete denial of service.
Details
The vulnerability exists in lib/core/mergeConfig.js at lines 98-101:
```javascript
utils.forEach(Object.keys({ ...config1, ...config2 }
Axios has a NO_PROXY Hostname Normalization Bypass that Leads to SSRF
Axios does not correctly handle hostname normalization when checking NO_PROXY rules.
Requests to loopback addresses like localhost. (with a trailing dot) or [::1] (IPv6 literal) skip NO_PROXY matching and go through the configured proxy.
This goes against what developers expect and lets attackers force requests through a proxy, even if NO_PROXY is set up to protect loopback or internal services.
According to RFC 1034 §3.1 and RFC 3986 §3.2.2, a hostname can have a trailing dot to show it is a fully qualified domain name (FQDN). At the DNS level, localhost. is the same as localhost.
However, Axios does a literal string comparison instead of normalizing hostnames before checking NO_PROXY. This causes requests like http://localhost.:8080/ and http://[::1]:8080/ to be incorrectly proxied.
This issue leads to the possibility of proxy bypass and SSRF vulnerabilities allowing attackers to reach sensitive loopback or internal services despite the configured protections.
---
PoC
``js
import http from "http";
import axios from "axios";
const proxyPort = 5300;
http.createServer((req, res) => {
console.log("[PROXY] Got:", req.method, req.url, "Host:", req.headers.host);
res.writeHead(200, { "Content-Type": "text/plain" });
res.end("proxied");
}).listen(proxyPort, () => console.log("Proxy", proxyPort));
process.env.HTTP_PROXY = http://127.0.0.1:${proxyPort};
process.env.NO_PROXY = "localhost,127.0.0.1,::1";
async function test(url) {
try {
await axios.get(url, { timeout: 2000 });
} catch {}
}
setTimeout(async () => {
console.log("\n[] Testing http://localhost.:8080/");
await test("http://localhost.:8080/"); // goes through proxy
console.log("\n[] Testing http://[::1]:8080/");
await test("http://[::1]:8080/"); // goes through proxy
}, 500);
`
Expected: Requests bypass the proxy (direct to loopback).
Actual: Proxy logs requests for localhost. and [::1].
---
Impact
Applications that rely on NO_PROXY=localhost,127.0.0.1,::1 for protecting loopback/internal access are vulnerable.
Attackers controlling request URLs can:
Force Axios to send local traffic through an attacker-controlled proxy.
Bypass SSRF mitigations relying on NO\_PROXY rules.
Potentially exfiltrate sensitive responses from internal services via the proxy.
---
Affected Versions
Confirmed on Axios 1.12.2 (latest at time of testing).
affects all versions that rely on Axios’ current NO_PROXY evaluation.
---
Remediation
Axios should normalize hostnames before evaluating NO_PROXY`, including:
Strip trailing dots from hostnames (per RFC 3986).
Normalize IPv6 literals by removing brackets for matching.
Next.js self-hosted applications vulnerable to DoS via Image Optimizer remotePatterns configuration
A DoS vulnerability exists in self-hosted Next.js applications that have remotePatterns configured for the Image Optimizer. The image optimization endpoint (/_next/image) loads external images entirely into memory without enforcing a maximum size limit, allowing an attacker to cause out-of-memory conditions by requesting optimization of arbitrarily large images. This vulnerability requires that remotePatterns is configured to allow image optimization from external domains and that the attacker can serve or control a large image on an allowed domain.
Strongly consider upgrading to 15.5.10 and 16.1.5 to reduce risk and prevent availability issues in Next applications.
Next.js Improper Middleware Redirect Handling Leads to SSRF
A vulnerability in Next.js Middleware has been fixed in v14.2.32 and v15.4.7. The issue occurred when request headers were directly passed into NextResponse.next(). In self-hosted applications, this could allow Server-Side Request Forgery (SSRF) if certain sensitive headers from the incoming request were reflected back into the response.
All users implementing custom middleware logic in self-hosted environments are strongly encouraged to upgrade and verify correct usage of the next() function.
More details at Vercel Changelog
Next.js Affected by Cache Key Confusion for Image Optimization API Routes
A vulnerability in Next.js Image Optimization has been fixed in v15.4.5 and v14.2.31. When images returned from API routes vary based on request headers (such as Cookie or Authorization), these responses could be incorrectly cached and served to unauthorized users due to a cache key confusion bug.
All users are encouraged to upgrade if they use API routes to serve images that depend on request headers and have image optimization enabled.
More details at Vercel Changelog
Next.js Content Injection Vulnerability for Image Optimization
A vulnerability in Next.js Image Optimization has been fixed in v15.4.5 and v14.2.31. The issue allowed attacker-controlled external image sources to trigger file downloads with arbitrary content and filenames under specific configurations. This behavior could be abused for phishing or malicious file delivery.
All users relying on images.domains or images.remotePatterns are encouraged to upgrade and verify that external image sources are strictly validated.
More details at Vercel Changelog
Authorization Bypass in Next.js Middleware
Impact
It is possible to bypass authorization checks within a Next.js application, if the authorization check occurs in middleware.
Patches
For Next.js 15.x, this issue is fixed in 15.2.3
For Next.js 14.x, this issue is fixed in 14.2.25
For Next.js 13.x, this issue is fixed in 13.5.9
For Next.js 12.x, this issue is fixed in 12.3.5
For Next.js 11.x, consult the below workaround.
_Note: Next.js deployments hosted on Vercel are automatically protected against this vulnerability.
axios Requests Vulnerable To Possible SSRF and Credential Leakage via Absolute URL
Summary
A previously reported issue in axios demonstrated that using protocol-relative URLs could lead to SSRF (Server-Side Request Forgery). Reference: axios/axios#6463
A similar problem that occurs when passing absolute URLs rather than protocol-relative URLs to axios has been identified. Even if baseURL is set, axios sends the request to the specified absolute URL, potentially causing SSRF and credential leakage. This issue impacts both server-side and client-side usage of axios.
##
Lodash has Prototype Pollution Vulnerability in `_.unset` and `_.omit` functions
Impact
Lodash versions 4.0.0 through 4.17.22 are vulnerable to prototype pollution in the _.unset and _.omit functions. An attacker can pass crafted paths which cause Lodash to delete methods from global prototypes.
The issue permits deletion of properties but does not allow overwriting their original behavior.
Patches
This issue is patched on 4.17.23.
Next.js Allows a Denial of Service (DoS) with Server Actions
Impact
A Denial of Service (DoS) attack allows attackers to construct requests that leaves requests to Server Actions hanging until the hosting provider cancels the function execution.
_Note: Next.js server is idle during that time and only keeps the connection open. CPU and memory footprint are low during that time._
Deployments without any protection against long running Server Action invocations are especially vulnerable. Hosting providers like Vercel or Netlify set a default maximum duration on function execution to reduce the risk of excessive billing.
This is the same issue as if the incoming HTTP request has an invalid Content-Length header or never closes. If the host has no other mitigations to those then this vulnerability is novel.
This vulnerability affects only Next.js deployments using Server Actions.
Patches
This vulnerability was resolved in Next.js 14.2.21, 15.1.2, and 13.5.8. We recommend that users upgrade to a safe version.
Workarounds
There are no official workarounds for this vulnerability.
Credits
Thanks to the PackDraw team for responsibly disclosing this vulnerability.
Next.js authorization bypass vulnerability
Impact
If a Next.js application is performing authorization in middleware based on pathname, it was possible for this authorization to be bypassed.
Patches
This issue was patched in Next.js 14.2.15 and later.
If your Next.js application is hosted on Vercel, this vulnerability has been automatically mitigated, regardless of Next.js version.
Workarounds
There are no official workarounds for this vulnerability.
Credits
We'd like to thank tyage (GMO
Denial of Service condition in Next.js image optimization
Impact
The image optimization feature of Next.js contained a vulnerability which allowed for a potential Denial of Service (DoS) condition which could lead to excessive CPU consumption.
Not affected:
The next.config.js file is configured with images.unoptimized set to true or images.loader set to a non-default value.
The Next.js application is hosted on Vercel.
Patches
This issue was fully patched in Next.js 14.2.7. We recommend that users upgrade to at least this version.
Workarounds
Ensure that the next.config.js file has either images.unoptimized, images.loader or images.loaderFile assigned.
Credits
Brandon Dahler (brandondahler), AWS
Dimitrios Vlastaras
Next.js Cache Poisoning
Impact
By sending a crafted HTTP request, it is possible to poison the cache of a non-dynamic server-side rendered route in the pages router (this does not affect the app router). When this crafted request is sent it could coerce Next.js to cache a route that is meant to not be cached and send a Cache-Control: s-maxage=1, stale-while-revalidate header which some upstream CDNs may cache as well.
To be potentially affected all of the following must apply:
Next.js between 13.5.1 and 14
Next.js Server-Side Request Forgery in Server Actions
Impact
A Server-Side Request Forgery (SSRF) vulnerability was identified in Next.js Server Actions by security researchers at Assetnote. If the Host header is modified, and the below conditions are also met, an attacker may be able to make requests that appear to be originating from the Next.js application server itself.
Prerequisites
Next.js (<14.1.1) is running in a self-hosted manner.
The Next.js application makes use of Server Actions.
The Server Action performs a redire
Axios Cross-Site Request Forgery Vulnerability
An issue discovered in Axios 0.8.1 through 1.5.1 inadvertently reveals the confidential XSRF-TOKEN stored in cookies by including it in the HTTP header X-XSRF-TOKEN for every request made to any host allowing attackers to view sensitive information.
jsonwebtoken's insecure implementation of key retrieval function could lead to Forgeable Public/Private Tokens from RSA to HMAC
Overview
Versions <=8.5.1 of jsonwebtoken library can be misconfigured so that passing a poorly implemented key retrieval function (referring to the secretOrPublicKey argument from the readme link) will result in incorrect verification of tokens. There is a possibility of using a different algorithm and key combination in verification than the one that was used to sign the tokens. Specifically, tokens signed with an asymmetric public key could be verified with a symmetric HS256 algorithm. This can lead to successful validation of forged tokens.
Am I affected?
You will be affected if your application is supporting usage of both symmetric key and asymmetric key in jwt.verify() implementation with the same key retrieval function.
How do I fix it?
Update to version 9.0.0.
Will the fix impact my users?
There is no impact for end users
