CVE Database

FileZen contains an OS command injection vulnerability. When FileZen Antivirus Check Option is enabled, a logged-in user may send a specially crafted HTTP request to execute an arbitrary OS command.

Summary A Reflected Cross-Site Scripting (XSS) vulnerability was discovered in the AI Playground's OAuth callback handler. The `error_description` query parameter was directly interpolated into an HTML script tag without proper escaping, allowing attackers to execute arbitrary JavaScript in the context of the victim's session. Root cause The OAuth callback handler in `site/ai-playground/src/server.ts` directly interpolated the `authError` value, sourced from the `error_description` query parameter, into an inline `<script>` tag. Impact An attacker could craft a malicious link that, when clicked by a victim, would: * Steal user chat message history - Access all LLM interactions stored in the user's session. * Access connected MCP Servers - Interact with any MCP servers connected to the victim's session (public or authenticated/private), potentially allowing the attacker to perform actions on the victim's behalf Mitigation: * PR:  https://github.com/cloudflare/agents/pull/841 https://github.com/cloudflare/agents/pull/841 * Agents-sdk users should upgrade to agents@0.3.10 * Developers using configureOAuthCallback with custom error handling in their own applications should ensure all user-controlled input is escaped before interpolation.

A vulnerability in the certificate validation logic may allow applications to accept untrusted or improperly validated server identities during TLS communication. An attacker in a privileged network position may be able to intercept or modify traffic if they can position themselves within the communication channel. Successful exploitation may compromise confidentiality, integrity, and availability of application data.

A permissive web security configuration may allow cross-origin restrictions enforced by modern browsers to be bypassed under specific circumstances. Exploitation requires the presence of an existing client-side injection vulnerability and user access to the affected web interface. Successful exploitation could allow unauthorized disclosure of sensitive information. Fixed in updated Omada Cloud Controller service versions deployed automatically by TP‑Link. No user action is required.

WWW::OAuth 1.000 and earlier for Perl uses the rand() function as the default source of entropy, which is not cryptographically secure, for cryptographic functions.

Improper restriction of operations within the bounds of a memory buffer in PCIe® Link could allow an attacker with access to a guest virtual machine to potentially perform a denial of service attack against the host resulting in loss of availability.

Solspace Freeform plugin for Craft CMS 5.x is a super flexible form-building tool. An authenticated, low-privilege user (able to create/edit forms) can inject arbitrary HTML/JS into the Craft Control Panel (CP) builder and integrations views. User-controlled form labels and integration metadata are rendered with dangerouslySetInnerHTML without sanitization, leading to stored XSS that executes when any admin views the builder/integration screens. This vulnerability is fixed in 5.14.7.

A cross-site scripting (XSS) vulnerability in the Item Kits function of OpenSourcePOS v3.4.1 allows attackers to execute arbitrary web scripts or HTML via injecting a crafted payload into the Item Name parameter.

ClamAV versions prior to 0.103.0-rc contain a vulnerability in function name processing through the ClamBC bytecode interpreter that allows attackers to manipulate bytecode function names. Attackers can exploit the weak input validation in function name encoding to potentially execute malicious bytecode or cause unexpected behavior in the ClamAV engine.

Centova Cast 3.2.12 contains a denial of service vulnerability that allows attackers to overwhelm the system by repeatedly calling the database export API endpoint. Attackers can trigger 100% CPU load by sending multiple concurrent requests to the /api.php endpoint with crafted parameters.

iNetTools for iOS 8.20 contains a denial of service vulnerability in the Whois feature that allows attackers to crash the application by manipulating input. Attackers can paste a specially crafted 98-character buffer into the Domain Name field to trigger an application crash.

SpotAuditor 5.3.2 contains a denial of service vulnerability in its Base64 decryption feature that allows attackers to crash the application by supplying an oversized buffer. Attackers can generate a malformed input file with 2000 repeated characters to trigger an application crash when pasted into the Base64 Encrypted Password field.

GHIA CamIP 1.2 for iOS contains a denial of service vulnerability in the password input field that allows attackers to crash the application. Attackers can paste a 33-character buffer of repeated characters into the password field to trigger an application crash on iOS devices.

DokuWiki 2018-04-22b contains a username enumeration vulnerability in its password reset functionality that allows attackers to identify valid user accounts. Attackers can submit different usernames to the password reset endpoint and distinguish between existing and non-existing accounts by analyzing the server's error response messages.

OwnCloud 8.1.8 contains a username enumeration vulnerability that allows remote attackers to discover user accounts by manipulating the share.php endpoint. Attackers can send crafted GET requests to /index.php/core/ajax/share.php with a wildcard search parameter to retrieve comprehensive user information.

SpotAuditor 5.3.2 contains a local buffer overflow vulnerability in the Base64 Encrypted Password tool that allows attackers to execute arbitrary code by crafting a malicious payload. Attackers can generate a specially crafted Base64 encoded payload to trigger a Structured Exception Handler (SEH) overwrite and execute shellcode on the vulnerable system.

PRO-7070 Hazır Profesyonel Web Sitesi version 1.0 contains an authentication bypass vulnerability in the administration panel login page. Attackers can bypass authentication by using '=' 'or' as both username and password to gain unauthorized access to the administrative interface.

Product Key Explorer 4.2.0.0 contains a denial of service vulnerability that allows local attackers to crash the application by overflowing the registration name input field. Attackers can create a specially crafted text file with repeated characters to trigger a buffer overflow when pasted into the registration name field, causing the application to crash.

Bullwark Momentum Series JAWS 1.0 contains a directory traversal vulnerability that allows unauthenticated attackers to access system files by manipulating HTTP request paths. Attackers can exploit the vulnerability by sending crafted GET requests with multiple '../' sequences to read sensitive files like /etc/passwd outside the web root directory.

FTP Commander Pro 8.03 contains a local stack overflow vulnerability that allows attackers to execute arbitrary code by overwriting the EIP register through a custom command input. Attackers can craft a malicious payload of 4108 bytes to overwrite memory and execute shellcode, demonstrating remote code execution potential.

AVS Audio Converter 9.1 contains a local buffer overflow vulnerability that allows local attackers to overwrite CPU registers by manipulating the 'Exit folder' input field. Attackers can craft a specially designed text file with 264 bytes of padding followed by register overwrite values to compromise the application and potentially execute arbitrary code.

SurfOffline Professional 2.2.0.103 contains a structured exception handler (SEH) overflow vulnerability that allows attackers to crash the application by manipulating the project name input. Attackers can generate a malicious payload of 382 'A' characters followed by specific byte sequences to trigger a denial of service condition and overwrite SEH registers.

FTP Navigator 8.03 contains a denial of service vulnerability that allows attackers to crash the application by overwriting Structured Exception Handler (SEH) with malicious input. Attackers can generate a payload of 4108 'A' characters followed by 4 'B' characters and 40 'C' characters to trigger a program crash when pasted into the custom command input.

XnConvert 1.82 contains a denial of service vulnerability in its registration code input field that allows attackers to crash the application. Attackers can generate a 9000-byte buffer of repeated characters and paste it into the registration code field to trigger an application crash.

Prime95 version 29.8 build 6 contains a buffer overflow vulnerability in the user ID input field that allows remote attackers to execute arbitrary code. Attackers can craft a malicious payload and paste it into the PrimeNet user ID and proxy host fields to trigger a bind shell on port 3110.

Thrive Smart Home 1.1 contains an SQL injection vulnerability in the checklogin.php endpoint that allows unauthenticated attackers to bypass authentication by manipulating the 'user' POST parameter. Attackers can inject malicious SQL code like ' or 1=1# to manipulate login queries and gain unauthorized access to the application.

RICOH Web Image Monitor 1.09 contains an HTML injection vulnerability in the address configuration CGI script that allows attackers to inject malicious HTML code. Attackers can exploit the entryNameIn and entryDisplayNameIn parameters to insert arbitrary HTML content, potentially enabling cross-site scripting attacks.

Heatmiser Netmonitor v3.03 contains an HTML injection vulnerability in the outputSetup.htm page that allows attackers to inject malicious HTML code through the outputtitle parameter. Attackers can craft specially formatted POST requests to the outputtitle parameter to execute arbitrary HTML and potentially manipulate the web interface's displayed content.

Heatmiser Netmonitor 3.03 contains a hardcoded credentials vulnerability in the networkSetup.htm page with predictable admin login credentials. Attackers can access the device by using the hard-coded username 'admin' and password 'admin' in the hidden form input fields.

FTP Navigator 8.03 contains a stack overflow vulnerability that allows attackers to execute arbitrary code by overwriting Structured Exception Handler (SEH) registers. Attackers can craft a malicious payload that triggers a buffer overflow when pasted into the Custom Command textbox, enabling remote code execution and launching the calculator as proof of concept.

E Learning Script 1.0 contains an authentication bypass vulnerability that allows attackers to access the dashboard without valid credentials by manipulating login parameters. Attackers can exploit the /login.php file by sending a specific payload '=''or' to bypass authentication and gain unauthorized access to the system.

Domain Quester Pro 6.02 contains a stack overflow vulnerability that allows remote attackers to execute arbitrary code by overwriting Structured Exception Handler (SEH) registers. Attackers can craft a malicious payload targeting the 'Domain Name Keywords' input field to trigger an access violation and execute a bind shell on port 9999.

AVS Audio Converter 9.1.2.600 contains a stack overflow vulnerability that allows attackers to execute arbitrary code by manipulating the output folder text input. Attackers can craft a malicious payload that overwrites stack memory and triggers a bind shell on port 9999 when the 'Browse' button is clicked.

Intego Personal Backup, a macOS backup utility that allows users to create scheduled backups and bootable system clones, contains a local privilege escalation vulnerability. Backup task definitions are stored in a location writable by non-privileged users while being processed with elevated privileges. By crafting a malicious serialized task file, a local attacker can trigger arbitrary file writes to sensitive system locations, leading to privilege escalation to root.

Intego Log Reporter, a macOS diagnostic utility bundled with Intego security products that collects system and application logs for support analysis, contains a local privilege escalation vulnerability. A root-executed diagnostic script creates and writes files in /tmp without enforcing secure directory handling, introducing a time-of-check to time-of-use (TOCTOU) race condition. A local unprivileged user can exploit a symlink-based race condition to cause arbitrary file writes to privileged system locations, resulting in privilege escalation to root.

Directus is a real-time API and App dashboard for managing SQL database content. Before 11.14.1, a timing-based user enumeration vulnerability exists in the password reset functionality. When an invalid reset_url parameter is provided, the response time differs by approximately 500ms between existing and non-existing users, enabling reliable user enumeration. This vulnerability is fixed in 11.14.1.

ntpd-rs is a full-featured implementation of the Network Time Protocol. Prior to 1.7.1, an attacker can remotely induce moderate increases (2-4 times above normal) in cpu usage. When having NTS enabled on an ntpd-rs server, an attacker can create malformed NTS packets that take significantly more effort for the server to respond to by requesting a large number of cookies. This can lead to degraded server performance even when a server could otherwise handle the load. This vulnerability is fixed in 1.7.1.

FastGPT is an AI Agent building platform. Due to the fact that FastGPT's web page acquisition nodes, HTTP nodes, etc. need to initiate data acquisition requests from the server, there are certain security issues. In addition to implementing internal network isolation in the deployment environment, this optimization has added stricter internal network address detection. This vulnerability is fixed in 4.14.7.

Scraparr is a Prometheus Exporter for various components of the *arr Suite. From 3.0.0-beta to before 3.0.2, when the Readarr integration was enabled, the exporter exposed the configured Readarr API key as the alias metric label value. Users were affected only if all of the following conditions are met, Readarr scraping feature was enabled and no alias configured, the exporter’s /metrics endpoint was accessible to external or unauthorized users, and the Readarr instance is externally accessible. If the /metrics endpoint was publicly accessible, the Readarr API key could have been disclosed via exported metrics data. This vulnerability is fixed in 3.0.2.

emp3r0r is a stealth-focused C2 designed by Linux users for Linux environments. Prior to 3.21.1, untrusted agent metadata (Transport, Hostname) is accepted during check-in and later interpolated into tmux shell command strings executed via /bin/sh -c. This enables command injection and remote code execution on the operator host. This vulnerability is fixed in 3.21.1.

Yoke is a Helm-inspired infrastructure-as-code (IaC) package deployer. In 0.19.0 and earlier, a vulnerability exists in the Air Traffic Controller (ATC) component of Yoke. It allows users with CR create/update permissions to execute arbitrary WASM code in the ATC controller context by injecting a malicious URL through the overrides.yoke.cd/flight annotation. The ATC controller downloads and executes the WASM module without proper URL validation, enabling attackers to create arbitrary Kubernetes resources or potentially escalate privileges to cluster-admin level.

Yoke is a Helm-inspired infrastructure-as-code (IaC) package deployer. In 0.19.0 and earlier, a vulnerability exists in the Air Traffic Controller (ATC) component of Yoke. The ATC webhook endpoints lack proper authentication mechanisms, allowing any pod within the cluster network to directly send AdmissionReview requests to the webhook, bypassing Kubernetes API Server authentication. This enables attackers to trigger WASM module execution in the ATC controller context without proper authorization.

grub-btrfs through 2026-01-31 (on Arch Linux and derivative distributions) allows initramfs OS command injection because it does not sanitize the $root parameter to resolve_device(). NOTE: a third party reports "exploitation may not be feasible under normal conditions and may depend on specific implementation details within resolve_device."

Airleader Master versions 6.381 and prior allow for file uploads without restriction to multiple webpages running maximum privileges. This could allow an unauthenticated user to potentially obtain remote code execution on the server.

lty628 aidigu v1.9.1 is vulnerable to Cross Site Scripting (XSS) exists in the /setting/ page where the "intro" field is not properly sanitized or escaped.

A flaw was found in Dropbear. When running in multi-user mode and authenticating users, the dropbear ssh server does the socket forwardings requested by the remote client as root, only switching to the logged-in user upon spawning a shell or performing some operations like reading the user's files. With the recent ability of also using unix domain sockets as the forwarding destination any user able to log in via ssh can connect to any unix socket with the root's credentials, bypassing both file system restrictions and any SO_PEERCRED / SO_PASSCRED checks performed by the peer.

AutoGPT is a platform that allows users to create, deploy, and manage continuous artificial intelligence agents that automate complex workflows. Prior to 0.6.48, an authenticated user could achieve Remote Code Execution (RCE) on the backend server by embedding a disabled block inside a graph. The BlockInstallationBlock — a development tool capable of writing and importing arbitrary Python code — was marked disabled=True, but graph validation did not enforce this flag. This allowed any authenticated user to bypass the restriction by including the block as a node in a graph, rather than calling the block's execution endpoint directly (which did enforce the flag). This vulnerability is fixed in 0.6.48.

navigation2 is a ROS 2 Navigation Framework and System. In 1.3.11 and earlier, a critical heap out-of-bounds write vulnerability exists in Nav2 AMCL's particle filter clustering logic. By publishing a single crafted geometry_msgs/PoseWithCovarianceStamped message with extreme covariance values to the /initialpose topic, an unauthenticated attacker on the same ROS 2 DDS domain can trigger a negative index write (set->clusters[-1]) into heap memory preceding the allocated buffer. In Release builds, the sole boundary check (assert) is compiled out, leaving zero runtime protection. This primitive allows controlled corruption of the heap chunk metadata(at least the size of the heap chunk where the set->clusters is in is controllable by the attacker), potentially leading to further exploitation. At minimum, it provides a reliable single-packet denial of service that kills localization and halts all navigation.

ClipBucket v5 is an open source video sharing platform. Prior to 5.5.3 - #45, in Clip Bucket V5, The Remote Play allows creating video entries that reference external video URLs without uploading the video files to the server. However, by specifying an internal network host in the video URL, an SSRF can be triggered, causing GET requests to be sent to internal servers. An attacker can exploit this to scan the internal network. Even a regular (non-privileged) user can carry out the attack.

XWiki Platform is a generic wiki platform offering runtime services for applications built on top of it. Prior to 17.9.0, 17.4.6, and 16.10.13, it's possible using comments to inject CSS that would transform the full wiki in a link area leading to a malicious page. This vulnerability is fixed in 17.9.0, 17.4.6, and 16.10.13.