CVE-2026-61539 Xinference: Remote code execution via unsafe `eval()` in Llama3 tool-call parsing

Xinference is an inference API for running open-source, speech, and multimodal models. In 2.5.0 and earlier, Xinference passes attacker-influenced Llama3 tool-call output to eval in xinference/model/llm/toolparsers/llama3toolparser.py and xinference/model/llm/utils.py. Requests to...

CVE-2026-18309

A flaw was found in GIMP. This vulnerability allows a remote attacker to execute arbitrary code on affected installations of GIMP. This occurs due to an integer overflow during the parsing of APNG Animated Portable Network Graphics files, caused by a lack of proper validation of user-supplied dat...

CVE-2026-18308

A flaw was found in GIMP. This vulnerability allows remote attackers to execute arbitrary code on affected installations. User interaction is required, as the target must open a malicious TIF Tagged Image File Format file or visit a malicious page. The issue stems from improper validation of...

CVE-2026-18306

A flaw was found in GIMP. This vulnerability, an integer overflow within the parsing of SGI Silicon Graphics Image files, allows a remote attacker to execute arbitrary code. Exploitation requires user interaction, where the target must open a specially crafted malicious SGI file. Successful...

CVE-2026-18305

A flaw was found in GIMP, an image manipulation program. This vulnerability allows a remote attacker to execute arbitrary code on an affected system. Exploitation requires user interaction, where the target must open a specially crafted TIF Tagged Image File Format file. The flaw occurs during TI...

CVE-2026-18304

A flaw was found in GIMP. This vulnerability allows a remote attacker to execute arbitrary code by tricking a user into opening a specially crafted TIF file. The issue arises from improper validation of user-supplied data during TIF file parsing, which can lead to an integer overflow before memor...

CVE-2026-18303

A flaw was found in GIMP. This vulnerability, a stack-based buffer overflow in the TIF file parsing component, allows a remote attacker to execute arbitrary code. User interaction is required, as the target must open a specially crafted malicious TIF file. The issue stems from insufficient...

CVE-2026-18302

A flaw was found in GIMP. This heap-based buffer overflow vulnerability in the TIF file parsing component allows remote attackers to execute arbitrary code. Successful exploitation requires user interaction, where a target must open a malicious TIF file. An attacker can leverage this to execute...

Cisco IOS XE Web UI - Command Injection

A vulnerability in the web UI component of Cisco IOS XE Software could allow an unauthenticated, remote attacker to execute arbitrary commands with root privileges on the underlying operating system. This vulnerability is due to improper input validation in the web UI. An attacker could exploit...

Apache OFBiz Directory Traversal - Remote Code Execution

Improper Limitation of a Pathname to a Restricted Directory 'Path Traversal' vulnerability in Apache OFBiz. This issue affects Apache OFBiz: before 18.12.13 id: CVE-2024-32113 info: name: Apache OFBiz Directory Traversal - Remote Code Execution author: DhiyaneshDK severity: high description: |...

Microsoft Exchange Server SSRF Vulnerability

This vulnerability is part of an attack chain that could allow remote code execution on Microsoft Exchange Server. The initial attack requires the ability to make an untrusted connection to Exchange server port 443. Other portions of the chain can be triggered if an attacker already has access or...

Confluence - Remote Code Execution

Confluence Server and Data Center is susceptible to an unauthenticated remote code execution vulnerability. id: CVE-2022-26134 info: name: Confluence - Remote Code Execution author: pdteam,jbertman severity: critical description: | Confluence Server and Data Center is susceptible to an...

CWP (Control Web Panel) < 0.9.8.1205 - Remote Code Execution

CWP Control Web Panel 0.9.8.1205 contains a remote code execution caused by shell metacharacters in the ttotal parameter in filemanager changePerm request, letting unauthenticated attackers execute code remotely, exploit requires knowledge of a valid non-root username. id: CVE-2025-48703 info:...

CVE-2026-18299

A flaw was found in GStreamer. This use-after-free vulnerability in the rtpsbcdepay element, specifically during the processing of Real-time Transport Protocol RTP payload elements, allows a remote attacker to execute arbitrary code. The issue arises from a lack of validation for an object's...

CVE-2026-18298

A flaw was found in GStreamer. This vulnerability, a heap-based buffer overflow during PNG file parsing, allows a remote attacker to execute arbitrary code. Successful exploitation requires user interaction, such as opening a malicious PNG file or visiting a specially crafted web page. The issue...

Livewire2025CVE

Livewire2025CVE Scanner Livewire CVE-2025-54068 Scanner automatizado de vulnerabilidades para detectar a CVE-2025-54068 em aplicações Laravel que usam Livewire v3.0.0-beta.1 até v3.6.3. Visão Geral Esta ferramenta escaneia uma lista de sites em busca de indicadores da vulnerabilidade crítica de...

CVE-2026-41449

UAC Unix-like Artifacts Collector versions prior to 3.3.0 contain a command injection vulnerability in the runcommand function that allows attackers to execute arbitrary commands by injecting shell metacharacters into untrusted data such as usernames, process names, or filenames. Attackers can...

CVE-2026-66786

A flaw was found in submariner. In cert-auth mode, the connection configuration is built using free-form strings from the Custom Resource Definition CRD without proper validation. A malicious cluster can exploit this by publishing a CableName that includes newlines and ipsec.conf directives. This...

CVE-2026-62675 Omnigent: Uploaded Agent Bundle Allows Authenticated Runner RCE via Python Callable Tools

Omnigent is an open-source AI agent framework and meta-harness for orchestrating coding agents. Prior to 0.3.0, multipart POST /v1/sessions accepts an authenticated user's agent bundle and omnigent/server/bundles.py validateagentbundle does not reject a tools..callable dotted Python path...

CVE-2026-62675 Omnigent: Uploaded Agent Bundle Allows Authenticated Runner RCE via Python Callable Tools

Omnigent is an open-source AI agent framework and meta-harness for orchestrating coding agents. Prior to 0.3.0, multipart POST /v1/sessions accepts an authenticated user's agent bundle and omnigent/server/bundles.py validateagentbundle does not reject a tools..callable dotted Python path...