CVE Vulnerability Catalog

An issue in libtiff 85f2ac8e0b01cb7db2bbecf4a3b891bdbef67938 allows an attacker to execute arbitrary code via the libtiff/tools/thumbnail.c: main() component.

Buffer Overflow vulnerability in gpac 31becc9e08b88e525a4a62013a4000de1c0f8fd9 allows an attacker to execute arbitrary code via the svgNameToImplementationName() function.

The WP Fastest Cache WordPress plugin before version 1.5.1 does not validate the Host header before using it to build URLs of asset files embedded in cached pages and does not include that header in the cache key, allowing unauthenticated attackers to poison cached pages with references to a server they control and have arbitrary JavaScript run for every subsequent visitor.

OpenEXR before versions 3.2.11, 3.3.13, and 3.4.14 contains a vulnerability in the exrmultiview utility that can write past a heap allocation when combining two attacker-supplied, individually valid scanline EXR files whose union dataWindow is not aligned to one view's channel subsampling. This leads to a heap out-of-bounds write.

OpenEXR versions 3.3.0 through 3.3.12 and 3.4.0 through 3.4.13 contain a heap buffer overflow in PyOpenEXR triggered by a channel-name key collision between literal and prefixed RGB channels. A crafted EXR file can cause an out-of-bounds write in a NumPy array during pixel decoding.

The NEService auto-upgrade process insecurely handles temporary files in the SonicWall NetExtender Linux client, allowing an attacker to manipulate file paths.

A Path traversal vulnerability in the SonicWall NetExtender Linux client file extractor component allows an attacker to write arbitrary files as root.

OpenEXR versions through 3.2.10, 3.3.0 through 3.3.12, and 3.4.0 through 3.4.13 contain a vulnerability in the OpenEXRUtil library that returns an out-of-bounds pointer from the SampleCountChannel::row() API for deep images with a non-zero dataWindow origin. This can lead to an out-of-bounds read, potentially causing a process crash or disclosure of memory data.

icalendar versions 7.1.0 through 7.1.3 contain a vulnerability in the Component equality method that performs O(2^n) work relative to nesting depth. An attacker can submit a small .ics file with deeply nested BEGIN:VEVENT blocks, potentially pinning a CPU core for a long time, causing denial of service.

Chainlit versions from 2.4.0rc0 to 2.12.0, when features.mcp.enabled is set to true, expose the POST /mcp endpoint without authentication. An attacker can control the URL and headers, allowing blind outbound requests to internal or external services, including cloud metadata endpoints, with arbitrary Authorization and Cookie headers. This issue is fixed in version 2.12.0.

A Content Security Policy bypass was addressed in AudioWorklet contexts in Safari and Apple operating systems. Processing maliciously crafted web content may bypass CSP.

Airbyte Platform incorrectly determines the workspace used for authorization from caller-supplied data rather than the resource identifier. This allows a member of any workspace to read source and destination configurations, trigger and cancel syncs, and delete connections, sources, and destinations belonging to workspaces they have no access to.

In Dradis Community Edition, the admin_required check is never applied because the constant Dradis::Pro is not defined in CE. This allows any authenticated non-admin user to create an AI provider pointing to an arbitrary HTTP/HTTPS address and reassign the built-in Roslin agent to use it, leading to server-side request forgery (SSRF) with the response reflected to the attacker.

Coroot's unauthenticated MCP OAuth dynamic client registration endpoint accepts any syntactically valid redirect URI without validation, allowing attackers to register clients pointing to attacker-controlled hosts. Attackers can send authorization URLs to signed-in users, capture their authorization codes, and exchange them for access tokens to hijack MCP sessions.

The python_repl tool in Amazon Strands Agents Tools before 0.8.5 has improper neutralization of input used for LLM prompting, allowing remote attackers to bypass the human consent gate and execute arbitrary Python code on the agent's host by forwarding non_interactive_mode as a keyword argument through the batch tool. This issue is fixed in version 0.8.5.

eml_parser before version 3.0.2 has a quadratic-time regex-based loop in the noparenthesis function when processing Received: headers with nested parentheses. A single header with 5000 nested parentheses causes about 1.3 seconds of CPU saturation per message, and doubling the nesting depth quadruples the running time. Attackers can submit small EML files that consume multiple seconds of processing time, causing latency and potential outages.

In sublinear-time-solver before 1.6.0 and consciousness-explorer before 1.1.2, the export_state, import_state, saveVectorToFile, and loadVectorFromFile tools pass attacker-controlled file paths to filesystem operations without restrictions. An attacker can read, write, or overwrite any file accessible to the server process, causing confidentiality and integrity loss and possible service disruption.

In Emacs TRAMP, a vulnerability exists where login arguments are concatenated without proper sanitization and passed to a local shell. A local attacker can exploit this by processing maliciously crafted filenames, potentially leading to arbitrary code execution.

Substance3D - Painter is affected by an out-of-bounds write vulnerability that could result in arbitrary code execution in the context of the current user. Exploitation requires user interaction, such as opening a malicious file.

Substance3D - Painter is affected by a Heap-based Buffer Overflow vulnerability that could result in arbitrary code execution in the context of the current user. Exploitation requires user interaction, such as opening a malicious file.