A command injection vulnerability exists in the bash activation script of virtualenv when the shell option 'promptvars' is off at activation time. The bash activator inserts the user-supplied --prompt value into PS1 without escaping special characters, allowing subsequent re-enabling of prompt expansion to execute arbitrary commands embedded in the prompt. This regression was introduced in version 21.14.9 and persists in 21.14.12. The vulnerability allows an attacker controlling the prompt value to execute commands as the user who activates the virtual environment.
CTFer.io's CTFd Terraform Provider manages CTFd resources as code. From 0.1.5 until 2.8.2, ctfd_team.password is stored as a non-sensitive value in Terraform state. A local or remote user with access to that state can read the team password without an additional decryption step. This issue is fixed in 2.8.2.
CVE-2026-87086 is a denial of service vulnerability in Tanium Provision caused by an unchecked return value leading to a NULL pointer dereference. This flaw can cause the affected software to crash or become unavailable. Tanium has released fixes in versions 1.17.150, 1.23.161, and 1.24.199 to address this issue.
Atlantis is a self-hosted golang application that listens for Terraform pull request events via webhooks. From version 0.4.0 until 0.47.1, Atlantis passes the pull-request-controlled workspace value from atlantis.yaml into Terraform command arguments that are joined with strings.Join(args, " ") and executed through sh -c. An authenticated contributor whose pull request is eligible for an Atlantis plan can include shell metacharacters in the workspace name, causing arbitrary commands to execute as the Atlantis service account without custom workflows, apply permission, or a malicious Terraform provider. External contributors can reach the flaw through fork pull requests only when --allow-fork-prs is enabled, while eligible same-repository branches do not require that option. The commands inherit the Atlantis process environment and can access cloud-provider, Terraform, and version-control credentials as well as files, state, and network resources available to the service. This issue is fixed in version 0.47.1.
systeminformation is a System and OS information library for node.js. Prior to 5.33.13, networkStats(ifaces) in lib/network.js on FreeBSD, OpenBSD, and NetBSD stores per-interface data in the plain-object _network cache and indexes it with the sanitized caller-controlled interface name. Passing proto causes the inherited Object.prototype value to be used instead of creating an own cache entry, so calcNetworkSpeed() writes rx_bytes, tx_bytes, rx_sec, tx_sec, ms, last_ms, and operstate onto Object.prototype. The polluted property names are fixed and their values come from interface statistics, but inherited-name collisions and enumerable keys can corrupt application branching, serialization, configuration merging, validation, or lookup behavior. This issue is fixed in version 5.33.13.
systeminformation is a System and OS information library for node.js. Prior to 5.33.14, the Linux networkStats(iface) implementation uses the caller-supplied interface value in an fs.existsSync() check and to build /sys/class/net/<iface> paths, including the operstate path, without verifying that it is a single interface-name segment or that the resolved path remains under /sys/class/net. An attacker who can control iface and create a directory containing symlinked statistics files can use parent-directory components and symlink traversal to make the API return the first line of a file readable by the Node.js process. The package does not expose a network endpoint, and the demonstrated scenario requires local filesystem setup. This issue is fixed in version 5.33.14.
A vulnerability in rustls versions prior to 0.23.46 and between 0.24.0-dev.1 and 0.24.0-dev.2 allows the TLS master secret or TLS 1.3 resumption PSK to be exposed when formatting a client session value with the Debug trait. This can lead to leakage of sensitive key material if a custom ClientSessionStore logs these debug outputs. Built-in diagnostics do not expose these secrets, and the session-store interface warns about the sensitivity of these values.
LiteLLM versions prior to 1.95.0 contain a vulnerability allowing an authenticated user with internal_user_viewer privileges to read arbitrary local files via a crafted vertex_ai_credentials value. This occurs because the request-body safety blocklist missed the alias vertex_ai_credentials, enabling file read access that can disclose sensitive environment variables and escalate privileges. The issue is fixed in version 1.95.0.
A critical vulnerability in better-auth versions from 1.4.0-beta.18 up to but not including 1.7.7 allows an unauthenticated attacker to use the OAuth state value as a Magic Link token to sign in as any user by knowing only their email address. This bypasses email verification and OAuth provider authentication, granting full session access. The flaw affects configurations using Magic Link with default or hashed token storage combined with social or Generic OAuth sign-in and database-backed OAuth state storage. The issue is fixed in version 1.7.7.
The msgpack-python C extension has a memory leak vulnerability when deserialization rejects a map key with strict_map_key validation enabled (default). This leak occurs because the decoded map value is not released on key validation failure, causing unbounded memory growth and potential denial of service. Additionally, the streaming Unpacker API may suffer from use-after-free issues due to stale parser-stack references. The issue affects msgpack versions prior to 1.2.3 and has been fixed in 1.2.3.
Google Guava versions from 4.0 up to but not including 33.7.2 contain a deserialization vulnerability in CompactHashMap, CompactHashSet, and MapMakerInternalMap. These classes read an attacker-controlled size value from a serialized stream without proper upper bound validation, leading to uncontrolled memory allocation. This can cause the Java Virtual Machine (JVM) to allocate multi-gigabyte arrays and crash with an OutOfMemoryError, even when the serialized data contains only a small number of actual elements.
dcmjs, a JavaScript library for DICOM manipulation used in web-based medical imaging viewers, contains an input validation bypass in its handling of DS (Decimal String) and IS (Integer String) value representations. The vulnerability allows malformed input strings to bypass sanitization and produce JavaScript Infinity or NaN values, or accept scientific notation in integer fields, violating DICOM standards. This can corrupt spatial geometry, image display, radiation dose calculations, and other clinical data in downstream applications. The issue is fixed in dcmjs version 0.52.0.
The ai-sdlc plugin MCP server through 0.31.2 contains an OS command injection vulnerability that allows attackers to execute commands by supplying shell metacharacters in the repo argument. The unvalidated repo value is interpolated into gh issue view or gh pr view commands run via execSync(), letting prompt-injected agents execute commands as the MCP server user.
SemanticMediaWiki versions from 2.5.0 up to but not including 7.3.1 contain a stored cross-site scripting (XSS) vulnerability. This flaw allows users with page-edit rights to inject malicious script via Code-typed property values. The vulnerability arises because attackers can embed HTML markup inside a valid JSON value, which is then rendered unescaped on the Special:Browse page, leading to JavaScript execution in the browsers of users viewing the page.
Semantic MediaWiki versions from 3.0.0 up to but not including 7.3.1 contain a cross-site scripting (XSS) vulnerability. This flaw arises because malicious semantic property values are rendered unescaped in autocomplete suggestions, allowing attackers to inject markup. Specifically, the jquery.autocomplete.js script inserts raw API values into HTML via container.html(), which can alter the dropdown content and potentially execute arbitrary scripts when a user opens the property-value autocomplete with an empty query. The vulnerability is fixed in version 7.3.1.
CVE-2026-103359 is a vulnerability in the PayDunya WooCommerce Payment Gateway WordPress plugin up to version 1.1.6. The plugin fails to verify the authenticity of incoming payment notifications properly, relying on a value derived from an empty secret before the gateway is configured. This flaw allows unauthenticated attackers to mark arbitrary WooCommerce orders as paid and completed.
A security vulnerability exists in DAMO Academy VideoLLaMA2 versions up to 1.1 in the load_pretrained_model function. The issue involves improper validation of an integrity check value due to manipulation of the _name_or_path argument. Remote exploitation is possible, and public exploit details have been disclosed. The project has been informed but has not yet responded or issued a fix.