| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| A vulnerability in MLflow versions <=3.10.1.dev0 allows unauthorized access to multipart upload (MPU) endpoints when the `--serve-artifacts` mode is enabled. The authorization logic does not enforce resource-level permission checks for `/mlflow-artifacts/mpu/*` endpoints, enabling attackers to overwrite artifacts belonging to other users. This can lead to unauthorized cross-user writes, model supply chain poisoning, and arbitrary code execution when compromised models are loaded. The issue is resolved in version 3.10.0. |
| FOSSBilling is a free, open-source billing and client management system. Prior to version 0.8.0, the password reset confirmation endpoint `/client/reset-password-confirm/:hash` is handled by a non-API controller and is not covered by FOSSBilling's rate limiter, which only applies to `/api/*` routes. This allows an attacker to probe the endpoint for valid reset tokens without any per-IP request limiting, attempt counting, or lockout mechanism. The endpoint acts as an oracle, returning a distinguishable response for valid versus invalid tokens (HTTP 200 vs HTTP 302 redirect). An attacker can submit unlimited token guesses to the password reset confirmation endpoint with no throttling applied. However, practical exploitability is significantly mitigated by the current token generation, which uses `hash('sha256', random_bytes(32))`, providing 256 bits of entropy. Tokens also expire after 15 minutes and are deleted after successful use. The same architectural gap applies to other controller-served auth routes, including `/staff/email/:hash` (admin password reset confirmation) and `/client/confirm-email/:hash` (email confirmation). Version 0.8.0 fixes the issue. Some workarounds are available. Configure a reverse proxy (e.g., Nginx, Apache, Cloudflare) to apply per-IP rate limiting to the `/client/reset-password-confirm/*` and `/staff/email/*` paths and/or use a WAF rule to limit request rates to these endpoints. |
| SmarterTools SmarterMail builds prior to 9560 contain a local file inclusion vulnerability in the /api/v1/report/summary/{type} API endpoint that allows authenticated users to read arbitrary .json files on the system. Attackers can exploit this vulnerability combined with weak encryption algorithms and hardcoded keys to decrypt and access stored passwords and 2FA secrets for all users. |
| PackagePersister.validate_tgz builds "tar -tf #{tgz} 2>&1" where tgz = File.join(release_dir, 'packages', "#{name}.tgz") and name = package_meta['name'] comes directly from release.MF inside the uploaded tarball. The string is passed to Bosh::Common::Exec.sh, which executes via %x{} — i.e., /bin/sh -c. No Shellwords.escape is applied. The Models::Package Sequel validation (VALID_ID = /^[-0-9A-Za-z_+.]+$/i) would reject the name, but in create_package (lines 74–79) the shell-out in save_package_source_blob runs before package.save, so validation fires too late.
Affected versions:
- BOSH: all versions prior to v282.1.12 (inclusive); fixed in v282.1.12 or later |
| SmarterTools SmarterMail builds prior to 9610 contain a cryptographic weakness in the file and email sharing endpoints that use DES-CBC encryption with keys and initialization vectors derived from System.Random seeded with insufficient entropy, reducing the seed space to approximately 19,000 possible values. An unauthenticated attacker can use the attachment download endpoint as an oracle to determine the seed in use and derive encryption keys and initialization vectors to forge sharing tokens for arbitrary emails, attachments, or file storage contents without prior access to the targeted content. |
| Missing input source validation in the tool authorization prompt in Kiro CLI before 1.28.0 allows a local attacker to execute arbitrary tools, including shell commands, without user approval by crafting content that is piped to kiro-cli via stdin.
We recommend you to upgrade to kiro-cli version 1.28.0 or later. |
| NVIDIA Transformers4Rec for Linux contains a vulnerability where an attacker could cause improper deserialization of untrusted data. A successful exploit of this vulnerability might lead to code execution, data tampering, and information disclosure. |
| Hermes WebUI contains an arbitrary file deletion vulnerability in the /api/session/delete endpoint that allows authenticated attackers to delete files outside the session directory by supplying an absolute path or path traversal payload in the session_id parameter. Attackers can exploit unvalidated session identifiers to construct paths that bypass the SESSION_DIR boundary and delete writable JSON files on the host system. |
| A weakness has been identified in PaddlePaddle FastDeploy up to 2.4.1. Affected by this issue is the function hash_features of the file fastdeploy/multimodal/hasher.py of the component MultimodalHasher. Executing a manipulation can lead to use of weak hash. The attack requires local access. A high complexity level is associated with this attack. The exploitation is known to be difficult. This patch is called 374945747652a8d32965591c0c01a00c88b7067f. Applying a patch is advised to resolve this issue. |
| A vulnerability was detected in zilliztech GPTCache up to 0.1.44. Affected by this issue is the function BufferedReader.peek of the file gptcache/processor/pre.py of the component Cache Key Handler. Performing a manipulation of the argument input_data["image"] results in use of weak hash. The attack must be initiated from a local position. The attack is considered to have high complexity. The exploitation is known to be difficult. The exploit is now public and may be used. The pull request to fix this issue awaits acceptance. |
| A vulnerability was found in mjperpinosa stumasy. The affected element is an unknown function of the file application/PHP/objects/updates/add_post.php. Performing a manipulation of the argument up_file_to_post results in unrestricted upload. The attack may be initiated remotely. The exploit has been made public and could be used. This product uses a rolling release model to deliver continuous updates. As a result, specific version information for affected or updated releases is not available. The project was informed of the problem early through an issue report but has not responded yet. |
| Tautulli is a Python based monitoring and tracking tool for Plex Media Server. Prior to version 2.17.1, a path traversal vulnerability in the cache deletion endpoint allows authenticated API access to delete directories outside the configured cache path. This can cause arbitrary data loss and service disruption. Version 2.17.1 fixes the issue. |
| LabF nfsAxe 3.7 Ping Client contains a buffer overflow vulnerability that allows local attackers to execute arbitrary code by supplying a malicious payload in the Host IP field. Attackers can craft a specially formatted input file with shellcode and overwrite the return address to execute calc.exe or other arbitrary commands. |
| AllPlayer 7.4 contains a local buffer overflow vulnerability in URL handling that allows attackers to overwrite structured exception handling pointers by supplying an excessively long URL string. Attackers can craft a malicious URL, paste it into the Open URL dialog, and trigger SEH-based code execution to run arbitrary commands with user privileges. |
| Care2x 2.7 contains multiple SQL injection vulnerabilities that allow unauthenticated attackers to execute arbitrary SQL commands by manipulating the ck_config cookie parameter. Attackers can inject malicious SQL through the ck_config cookie in multiple endpoints including login.php, indexframe.php, and various module files to extract sensitive database information without authentication. |
| Dell BSAFE SSL-J contains an allocation of resources without limits or throttling vulnerability. An unauthenticated remote attacker could potentially exploit this vulnerability, leading to a Denial of Service (DoS). |
| An authorization flaw existed in the MISP Event Template Importer overwrite workflow. When importing an event template in overwrite mode, the application checked whether a matching template already existed but did not verify that the importing user belonged to the organization that owned the existing template. As a result, an authenticated user with access to the template import functionality could forcibly overwrite an event template owned by another organization.
Successful exploitation could allow unauthorized modification of another organization’s event template, potentially altering template structure, attributes, or metadata used for subsequent event creation or sharing workflows. Site administrators are not affected by this restriction, as they are explicitly allowed to overwrite templates across organizations.
The issue was fixed by enforcing an ownership check before overwrite: non-site-admin users may only overwrite templates owned by their own organization. |
| A URL validation flaw in the MISP dashboard button widget allowed a crafted relative-looking URL to be accepted as a local path while being interpreted by browsers as an external URL. The validation rejected URLs containing an explicit scheme, host, or user component, but did not reject paths beginning with a slash followed by a backslash, such as /\example.com. Some browsers normalize backslashes in URLs as forward slashes, which can turn this into a scheme-relative external navigation target. In addition, the generated href concatenated the reconstructed URL with the original URL, increasing the possibility of unsafe or malformed link generation.
An attacker able to configure or influence a dashboard button URL could craft a button that appears to point inside the application but redirects users to an attacker-controlled site when clicked. This could be used for phishing, credential theft, or social engineering. The patch fixes the issue by rejecting empty paths and paths starting with /\, and by emitting only the reconstructed validated URL in the anchor href. |
| An open redirect vulnerability existed in MISP UsersController::routeafterlogin() because the value stored in the pre_login_requested_url session key was used as the post-login redirect destination without sufficiently enforcing that it was a local application path.
An unauthenticated remote attacker could craft a link that causes a victim to visit a trusted MISP instance and, after successful authentication, be redirected to an attacker-controlled external URL. This could be abused to increase the credibility of phishing attacks, redirect users to counterfeit login pages, or deliver attacker-controlled content from an untrusted domain. CWE-601 describes this weakness as accepting user-controlled input that specifies an external link and using it in a redirect, with phishing as a common consequence.
The patch mitigates the issue by decoding and parsing the URL, rejecting URLs with a scheme, host, user component, missing or non-local path, and protocol-relative forms such as //example.com and /\example.com. |
| A flaw was found in Red Hat Quay's handling of resumable container image layer uploads. The upload process stores intermediate data in the database using a format that, if tampered with, could allow an attacker to execute arbitrary code on the Quay server. |
