A malicious npm package named spectral-core impersonates a legitimate project and executes code during installation to steal sensitive environment variables and AWS EC2 metadata. It collects CI/CD secrets, cloud credentials, network reconnaissance data, and sends this information to a hardcoded external server. The package contains no functional library code and uses a binding.gyp file solely to run the malicious script at install time.
The npm package @dmangal/curse-dependent-ecto contains malicious code embedded in its binding.gyp file. During installation, it executes a node script that searches for files containing the marker 'HTB{' and reads sensitive files and environment variables. It then exfiltrates this data, including CI secrets and API tokens, to an attacker-controlled endpoint. The package uses an implausibly high version number consistent with a dependency confusion attack. Installing this package in developer or CI environments risks leaking confidential information to attackers.
The b20miner npm package version 0.3.1 contains malicious code that compromises any computer on which it is installed or running. The compromise is severe enough that all secrets and keys stored on the affected system should be rotated immediately from a different, uncompromised computer. Removal of the package alone may not eliminate all malicious software introduced by its installation.
The systeminformation package version 5.33.15 on Windows is vulnerable to command injection via the services() function. The vulnerability arises because the sanitizer removes ASCII quotes but not Unicode curly quotes, allowing crafted input to break out of a PowerShell filter string and execute arbitrary commands with the Node.js process permissions. This can lead to unauthorized command execution if attacker-controlled input reaches services(). The issue is fixed in version 5.33.16.
A vulnerability in @backstage/plugin-catalog-backend allows bypassing catalog location rules by using encoded path segments in URLs. This causes the system to accept entities from untrusted sources as if they were from trusted locations, potentially leading to unauthorized ingestion of entity content. The issue is fixed in versions 3.9.2 and 4.0.2. No complete workaround exists other than restricting location registration permissions.
The Waline login endpoint in @waline/vercel before version 1.41.0 accepts a non-string email value, which leads to a MongoDB query filter that matches all users. This causes the login process to verify the submitted password against the first-registered user (the administrator) without requiring the correct email. This vulnerability enables partial authentication bypass, user enumeration, and credential-stuffing amplification against the administrator account.
Seroval versions 1.6.0 through 1.6.10 contain a vulnerability where typed arrays that have an own constructor property set from untrusted data can lead to code injection during serialization. This occurs because the constructor name is read from a mutable property on the typed array rather than its internal type, allowing arbitrary code to be serialized. The issue is fixed in version 1.6.11. The vulnerability requires unusual conditions where properties from untrusted objects are copied onto typed arrays.
Seroval versions before 1.0.0 and from 1.5.5 up to but not including 1.6.10 improperly serialize RegExp sources containing the '<' character without escaping it. This can allow an attacker to inject script code if the RegExp is built from untrusted input and the serialized output is placed inside an HTML <script> element. The vulnerability is fixed in version 1.6.10 by changing the serialization method to use the RegExp constructor with an escaped string source.
The Seroval library versions prior to 1.6.8 contain a denial-of-service vulnerability in the fromJSON and fromCrossJSON deserialization functions. These functions trust node-supplied array lengths and sequence doneAt indices without validation, allowing crafted inputs to cause infinite loops or prolonged blocking of the event loop. This can be triggered by unauthenticated requests passing untrusted input, resulting in service unavailability. The issue is fixed in version 1.6.8 by enforcing strict validation of array elements and sequence indices, throwing errors on malformed input.
Seroval versions from 1.4.1 up to but not including 1.6.8 contain a denial of service vulnerability in the fromJSON and fromCrossJSON functions. These functions fail to validate that an array node's element list is actually an array, allowing crafted payloads to control memory allocation and processing loops. This can cause the service to become unavailable due to excessive memory use or prolonged thread blocking. The issue bypasses a previous fix and affects any framework using these functions to decode untrusted request bodies before validation. The vulnerability is fixed in version 1.6.8.
A deserialization vulnerability in seroval versions prior to 1.6.8 allows attacker-controlled callables to be invoked during deserialization. This occurs because the fromJSON function calls on() on the source of an async iterator without verifying that the source is a legitimate seroval stream. Crafted payloads can exploit this to invoke arbitrary plugin functions during deserialization. The issue is fixed in version 1.6.8 by enforcing instanceof checks on source values.
A vulnerability in mem0ai's session key construction allows cross-tenant data leakage and corruption. The function _build_session_scope() concatenates user_id, agent_id, and run_id with unescaped '&' and '=' delimiters, enabling an attacker controlling only run_id or agent_id to craft values that collide with another tenant's session key. This leads to unauthorized reading and overwriting of another tenant's conversation history. The flaw affects both Python and TypeScript SDKs and is fixed in versions 2.0.18 and 3.1.6.
The systeminformation Node.js library version 5.33.13 on Linux contains a path traversal vulnerability in its networkStats(iface) function. This function constructs file paths under /sys/class/net/ from user-supplied interface names without validating that the resolved path remains within the expected directory. An attacker able to create symlinks and control the iface parameter can cause the function to read and return the first line of arbitrary files accessible to the Node process, including files with restricted permissions. The vulnerability is fixed in version 5.33.14. The library itself does not expose a network interface, and no remote exploit scenario has been demonstrated.
The systeminformation library version prior to 5.33.13 contains a prototype pollution vulnerability in the networkStats() function. This occurs when the interface name '__proto__' is passed, causing writes to Object.prototype and polluting all objects in the Node.js process. The vulnerability affects applications running on FreeBSD, OpenBSD, or NetBSD that pass untrusted input to networkStats(). The impact includes unexpected properties appearing on all objects, which can alter application logic and enumeration behavior. No remote code execution or attacker-controlled values are known. A fix is available in version 5.33.13.
The graphql-scalars package's PhoneNumber scalar is vulnerable to a regular expression denial of service (ReDoS) due to an exponential backtracking flaw introduced in version 1.24.1. This flaw allows specially crafted inputs of digits followed by a failing character to stall the JavaScript thread for seconds, with processing time doubling for each additional digit. The vulnerability affects all versions from 1.24.1 up to and including 2.0.0. The regex runs during GraphQL document validation and variable coercion, before any authentication or authorization, enabling unauthenticated requests to cause denial of service. A fix is available in version 2.0.1.
i18next-http-backend versions prior to 4.0.4 contain a vulnerability where the language parameter (lng) used in URL query strings is not properly sanitized for the characters '&' and '='. This allows an attacker controlling the language value to inject or override query parameters in requests sent by the backend, potentially manipulating the translation API requests. The vulnerability arises because the URL segment validation blocks certain characters but does not reject '&' or '=', and the language segments are inserted into URLs without encoding. A fix is available in version 4.0.4 that addresses this issue.
A quadratic-time denial of service vulnerability exists in re2js when using the opt-in LOOKBEHINDS flag with repeated-matching APIs such as matcher().find(), matchAll(), split(), or replaceAll(). This flaw causes the regex engine to restart scanning from the beginning of the input on every find call, resulting in quadratic time complexity and severe CPU consumption on large inputs. The issue affects re2js versions from 2.2.0 up to but not including 2.8.7 and is fixed in version 2.8.7.
A vulnerability in i18next-http-backend versions before 4.0.3 allows an attacker controlling the namespace (ns) value to cause the application to load translation JSON from an attacker-chosen origin. This occurs when the loadPath or addPath template places the namespace value immediately after a leading slash, resulting in a protocol-relative URL starting with //. This can lead to content injection into the UI and potential cross-site scripting (XSS) if translations are rendered as HTML. The issue is fixed in version 4.0.3.
happy-dom versions up to 20.8.9 perform real server-side HTTP requests when parsing untrusted HTML containing <iframe> or <link rel="stylesheet"> elements. This default behavior can be exploited to trigger Server-Side Request Forgery (SSRF), allowing attackers to make arbitrary HTTP requests from the server. The vulnerability can lead to cloud metadata access, internal network scanning, and data exfiltration. A fix is available in version 21.0.0 that disables external resource loading by default.
Vue I18n versions before 11.4.14 and certain 12.0.0-alpha versions have a vulnerability where the escapeParameter feature does not properly escape named interpolation values when a translation call passes both a list and named values. This can lead to unescaped HTML markup being inserted, potentially enabling DOM-based XSS if untrusted named values are rendered as HTML. The issue is fixed in versions 11.4.14 and 12.0.0-alpha.5.