53547 matches found
CVE-2026-54027
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.4-rc1, the POST /api/files/images endpoint allows any authenticated user to upload files into any agent's toolresources e.g., context, executecode without verifying ownership or EDIT permission on the target...
CVE-2026-55092
Trivy before 0.71.1 vulnerable to path traversal when downloading OCI artifacts: the org.opencontainers.image.title annotation from the artifact manifest is used as the destination filename without validation, allowing writing layer content to arbitrary locations on the host filesystem. Impact is...
CVE-2026-48945
The K2 article gallery upload path accepts a zip/tar archive, extracts it under /media/k2/galleries//, and only renames image files gif/jpg/jpeg/png/webp to safe names — non-image files including .php are extracted as-is and remain executable via direct HTTP access...
CVE-2026-48942
K2 ≤ 2.26 renders the k2users.image column directly into HTML src attributes via two distinct templates, in both cases without HTML escaping...
CVE-2026-54025 LibreChat: Stored XSS via unescaped image alt text in markdown artifact preview
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.4-rc1, there is a vulnerability in LibreChat's markdown artifact preview pipeline. The marked library v15.0.12 does not HTML-escape double-quote characters in image alt text when a custom renderer falls throu...
CVE-2026-54025
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.4-rc1, there is a vulnerability in LibreChat's markdown artifact preview pipeline. The marked library v15.0.12 does not HTML-escape double-quote characters in image alt text when a custom renderer falls throu...
EUVD-2026-39463
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.4-rc1, there is a vulnerability in LibreChat's markdown artifact preview pipeline. The marked library v15.0.12 does not HTML-escape double-quote characters in image alt text when a custom renderer falls throu...
CVE-2026-54025
LibreChat suffers a stored XSS in its Markdown artifact preview prior to version 0.8.4-rc1. The vulnerability arises because lib re uses marked v15.0.12 to render image alt text without HTML-escaping double quotes when the custom image renderer defers to the default renderer. LibreChat’s generate...
CVE-2026-54025 LibreChat: Stored XSS via unescaped image alt text in markdown artifact preview
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.4-rc1, there is a vulnerability in LibreChat's markdown artifact preview pipeline. The marked library v15.0.12 does not HTML-escape double-quote characters in image alt text when a custom renderer falls throu...
CVE-2026-54027
Vulnerability (CVE-2026-54027): LibreChat prior to 0.8.4-rc1 allows authenticated users to upload files via POST /api/files/images into any agent’s tool_resources (e.g., context, execute_code) without ownership/EDIT checks. A permission check was added to POST /api/files, but the image upload rou...
CVE-2026-54027 LibreChat: Image Upload Route Bypasses Agent Permission Check — Incomplete Fix for File Upload Authorization
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.4-rc1, the POST /api/files/images endpoint allows any authenticated user to upload files into any agent's toolresources e.g., context, executecode without verifying ownership or EDIT permission on the target...
CVE-2026-54027
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.4-rc1, the POST /api/files/images endpoint allows any authenticated user to upload files into any agent's toolresources e.g., context, executecode without verifying ownership or EDIT permission on the target...
CVE-2026-54027 LibreChat: Image Upload Route Bypasses Agent Permission Check — Incomplete Fix for File Upload Authorization
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.4-rc1, the POST /api/files/images endpoint allows any authenticated user to upload files into any agent's toolresources e.g., context, executecode without verifying ownership or EDIT permission on the target...
EUVD-2026-39447
The K2 article gallery upload path accepts a zip/tar archive, extracts it under /media/k2/galleries//, and only renames image files gif/jpg/jpeg/png/webp to safe names — non-image files including .php are extracted as-is and remain executable via direct HTTP access...
CVE-2026-48945
The CVE describes a vulnerability in the K2 Joomla extension (getk2.com) where the article gallery upload path accepts a zip/tar archive and extracts it to /media/k2/galleries//. The extractor renames image files (gif/jpg/jpeg/png/webp) to safe names, but non-image files (including .php) are extr...
CVE-2026-48945
The K2 article gallery upload path accepts a zip/tar archive, extracts it under /media/k2/galleries//, and only renames image files gif/jpg/jpeg/png/webp to safe names — non-image files including .php are extracted as-is and remain executable via direct HTTP access...
EUVD-2026-39439
K2 ≤ 2.26 renders the k2users.image column directly into HTML src attributes via two distinct templates, in both cases without HTML escaping...
CVE-2026-48942
Affected software: K2 extension for Joomla (getk2.com), version constraint listed as K2 ≤ 2.26. Vulnerability: two templates render the database column __#k2_users.image directly into HTML src attributes without HTML escaping, revealing a stored-XSS risk. Root cause: lack of escaping when injecti...
CVE-2026-48942
K2 ≤ 2.26 renders the k2users.image column directly into HTML src attributes via two distinct templates, in both cases without HTML escaping...
CVE-2026-48942 Joomla Extension - getk2.org - Stored-XSS in K2 extension for Joomla < 2.26
K2 ≤ 2.26 renders the k2users.image column directly into HTML src attributes via two distinct templates, in both cases without HTML escaping...