The Acadlix LMS plugin for WordPress up to version 1.1.9 contains a path traversal vulnerability in the post_delete_assessment_file_by_id function. This flaw allows authenticated users with subscriber-level access or higher to delete arbitrary files on the server due to insufficient file path validation. The plugin's permission check relies solely on a generic REST nonce exposed publicly, effectively allowing unauthenticated attackers to exploit this issue. Deleting critical files such as wp-config.php can lead to remote code execution. The vulnerability has a high CVSS score of 9.1, indicating critical severity.
The SimpleForm – Form Maker to Chat Apps with Quiz, Poll and Multi Lead Notifications WordPress plugin up to version 3.10.2 has an arbitrary file upload vulnerability. This flaw allows unauthenticated attackers to upload potentially executable files by bypassing file extension and MIME type validation using a trailing-dot filename. The vulnerability arises from missing server-side file key allowlist and improper sanitization in the handle_upload function. The only access control is a nonce embedded in public forms, which is easily accessible to any visitor. This can lead to remote code execution on affected installations.
The Project Manager – AI Powered plugin for WordPress up to version 4.1.1 contains a critical vulnerability that allows unauthenticated remote code execution. This occurs because the 'id' parameter is unsanitized and used directly in a dynamic PHP function call without validation, bypassing authorization checks. The flaw enables attackers to execute arbitrary code on the server, posing a severe risk to confidentiality, integrity, and availability.
The Shield Security plugin for WordPress, used for bot blocking, brute-force login protection, and file scanning, contains a stored cross-site scripting (XSS) vulnerability in the 'REQUEST_URI' parameter. This affects all versions up to and including 22.3.2. The vulnerability arises because the hidden 'path' column in ForTraffic.php is rendered using innerHTML without proper escaping, allowing injection of arbitrary scripts that execute when a user accesses the affected page. A partial fix was applied in version 18.5.8 but did not address the hidden 'path' field, leaving the issue exploitable through version 22.2.6. The CVSS 3.1 score is 7.2, indicating high severity.
A vulnerability in the Python package 'fickling' version 0.1.12 allows arbitrary code execution via a crafted pickle containing a STACK_GLOBAL operand with a dotted __qualname__. The flaw arises because the pickle's find_class resolves dotted attribute paths to arbitrary attributes rather than restricting to module-level functions, bypassing the security checks in fickling.load(). This enables execution of the optparse.Values.read_file method, which executes code from a file. The issue is distinct from other module-origin trust weaknesses and remains unpatched as of the latest vendor commit in October 2026.
A path traversal vulnerability in the Signal K server's resources API allows unauthenticated or low-privileged users to read, write, and delete files outside the intended storage directory. This includes reading the security.json file containing the JWT signing secret, enabling attackers to forge admin JWT tokens and gain administrative access. With admin privileges, attackers can create providers that execute arbitrary shell commands, resulting in remote code execution with server privileges. The vulnerability affects all signalk-server versions prior to 2.34.0 and resources-provider versions prior to 1.5.3. The issue is patched in versions 2.34.0 and 1.5.3.
AdminNeo versions prior to 5.5.1 contain a vulnerability in their SQLite interface that allows authenticated users to execute arbitrary code. The application blocks ATTACH queries to prevent arbitrary file creation, but does not block the VACUUM INTO command, which can write the entire database to an arbitrary file path. An attacker can exploit this by injecting PHP code into a table name and then using VACUUM INTO to write a web shell to the web root, enabling remote command execution.
CVE-2026-61725 is a path traversal vulnerability in hasenbanck sevenz-rust2, a Rust-native 7z compressor and decompressor. Versions prior to 0.21.1 allow an attacker to craft a malicious 7z archive that, when extracted, can write or overwrite files outside the intended destination directory. This occurs because the extraction code does not normalize or verify the archive entry paths, enabling directory traversal or absolute path overwrites. Such unauthorized file writes can lead to code execution or persistence if critical configuration or startup files are overwritten. The issue is fixed in version 0.21.1.
CVE-2026-63184 affects the christianhelle refitter tool, a C# REST API client generator using the Refit library. Versions prior to 2.1.1 invoke NJsonSchema with unsafe defaults for $ref resolution, allowing attacker-controlled OpenAPI descriptions to cause remote HTTP(S) requests and inline untrusted schema content. This leads to server-side request forgery, remote file inclusion, and local file inclusion vulnerabilities, permitting arbitrary local file disclosure. The issue does not lead to code execution due to runtime URL setting and string escaping. The vulnerability is fixed in version 2.1.1.
Trail of Bits fickling versions up to 0.1.12 contain a protection mechanism failure where the _process_import() method trusts all callables from standard library modules not explicitly denylisted. This leads to unsafe code execution or file writes when malicious pickle payloads invoke such callables, bypassing the check_safety() gate. The flaw affects Python 3.7 through 3.12 due to reliance on module origin for trust decisions. The vendor has added the demonstrated risky modules to the denylist on the master branch but no official release patch is yet available.
Tesseract OCR versions prior to 5.5.4 contain a vulnerability in the ReadAdaptedClass function in src/classify/adaptive.cpp. This function raw-reads an ADAPT_CLASS_STRUCT from a pre-adapted-template .a file without properly neutralizing certain pointer fields before destructor execution. When the legacy engine is used with classify_use_pre_adapted_templates enabled, an attacker able to write the output-basename .a file can cause memory corruption and denial of service by passing crafted pointers to the destructor and delete operator. This issue is fixed in version 5.5.4.
Metalama is a Roslyn-based open-source patterns and architecture toolkit for C#. Prior to 2025.1.18, 2026.0.25 and 2026.1.19, Metalama.Framework and Metalama.Tool load and execute compile-time assemblies, auxiliary executable tools and embedded helper assemblies from predictable temporary directories. On `Linux` and `macOS`, the default `Path.GetTempPath()` location is world-writable, and existing artifacts are trusted based on sentinel or `.completed` marker files rather than integrity verification. A local unprivileged user on a shared machine or CI/build agent can pre-create the deterministic directory structure and plant a malicious DLL with the matching sentinel. When a higher-privileged developer or build account subsequently builds, Metalama skips regeneration and executes the planted artifact in the compiler, IDE or MSBuild process as the building user. This permits cross-user privilege escalation, reading source and secrets, supply-chain tampering with build output, or disrupting builds. Default Windows per-user temporary directories are protected by ACLs and are not affected. Deployer-selected insecure cache locations are outside the vulnerability's scope. This issue is fixed in versions 2025.1.18, 2026.0.25 and 2026.1.19.
Pi-hole is a DNS sinkhole that protects devices from unwanted content without installing any client-side software. From 6.4.2 until 6.5, the root-run advanced/Scripts/updatecheck.sh cron job applies truncate -s 0 and chmod 644 to /etc/pihole/versions while following symbolic links in a directory owned by the pihole user. An attacker executing as pihole can replace that path with a symlink so the scheduled root operations truncate or create an arbitrary root-owned file, or race the operations so chmod makes a sensitive target world-readable without truncating it. This issue is fixed in version 6.5.
CVE-2026-119335 is an OS command injection vulnerability in pi-hole versions prior to 6.5. It arises from improper validation of the pihole-owned PID file, allowing an attacker with pihole user privileges to craft a malicious PID file that leads to command execution as root. This bypasses previous mitigations related to PID file validation. The issue is fixed in pi-hole version 6.5.
CVE-2026-119333 is a high-severity vulnerability in draw.io Desktop (an Electron app) prior to version 31.5.2. The Windows application launched system executables powershell and attrib by bare name, allowing an attacker to place malicious executables in the current working directory. When a user opened a .drawio file from a directory containing attacker-controlled files, this could lead to execution of malicious code. The issue is fixed in version 31.5.2.
CVE-2026-116250 is a time-of-check to time-of-use (TOCTOU) race condition vulnerability in pgAdmin 4 server mode before version 9.19. It arises when the application validates a user-supplied file path within the storage directory but subsequently opens the file without securely handling symbolic links or other special file types. This allows an attacker with write access to the storage backend to swap in symbolic links, hard links, or FIFOs to redirect file operations outside the intended directory, potentially leading to unauthorized file read, write, or execution as the pgAdmin OS user. The vulnerability was fixed in version 9.19 by securely opening files component-by-component with O_NOFOLLOW and passing open descriptors to utilities.
A vulnerability in the gettext family of functions in GNU C Library 2.0 and later allows attackers to exploit the LANGUAGE environment variable to load message catalogs from arbitrary filesystem locations. This can lead to program crashes or arbitrary code execution if a crafted .mo file is used. The issue arises because LANGUAGE entries containing '/' or '..' are accepted unless the program is setuid or setgid (AT_SECURE). Exploitation requires the attacker to control the environment and place malicious files in readable locations, and to select a locale other than "C", "C.<encoding>", or "POSIX".
Mythic versions 3.2.16 through 3.4.0.61 have a stored cross-site scripting (XSS) vulnerability. Attackers who control target endpoint files can embed malicious HTML in files named with a .pdf extension. When an operator previews such a file in the Mythic UI, the malicious content executes JavaScript in the same origin, allowing theft of localStorage JWT tokens. These tokens grant full GraphQL and REST API access.
Flowise versions through 3.1.4 have a sandbox escape vulnerability that allows authenticated users to execute arbitrary OS commands. This occurs when the ALLOW_BUILTIN_DEP=true setting is enabled, enabling abuse of allowlisted langchain and mammoth modules. Attackers can submit JavaScript code to the /api/v1/node-custom-function endpoint, which writes a file into the mammoth package via LocalFileStore and requires it, thereby escaping the vm2 sandbox as the Flowise process user.