CVE-2026-105148 is a server-side request forgery (SSRF) vulnerability in SciPhi-AI R2R versions 3.6.0 through 3.6.6. It affects the Retrieval Completion API Endpoint component, specifically the file py/shared/abstractions/llm.py. The vulnerability arises from manipulation of the argument generation_config.api_base, allowing remote attackers to induce the server to make unauthorized requests. Exploit code is publicly available, but no vendor response or patch has been provided.
The MPG WordPress plugin before 4.2.3 does not validate that the dataset source supplied when importing a project is a remote URL before treating it as a local filesystem path and copying that file into a publicly accessible uploads folder. This makes it possible for users with the Editor role and above to read the contents of arbitrary files on the server, with the copied file then retrievable by unauthenticated visitors.
Vibe-Trading AI contains two related path traversal vulnerabilities in its file-read tools, allowing unauthenticated users to read arbitrary files within the container's root home (/root) and application directory (/app). The safe_user_path() function incorrectly permits any path under these directories, exposing sensitive files such as SSH keys, AWS credentials, and environment variables. Additionally, the read_document() function lacks sandboxing, enabling reading of any file accessible to the FastAPI process, including critical system files like /etc/shadow. The container runs as root, exacerbating the risk. These vulnerabilities can leak sensitive file contents via error messages when the files are parsed as trade journals. A patch is available to address both issues.
CVE-2026-104120 is a server-side request forgery (SSRF) vulnerability in the fetch_url function of the mcp_server_fetch/server.py file within the modelcontextprotocol mcp-server-fetch and mcp-server-everything components up to version 2026.6.4. This vulnerability allows remote attackers to manipulate URL or path arguments to induce the server to make unintended requests. The vulnerability has been publicly disclosed, and a fix is pending acceptance of a pull request.
Red Hat Satellite 6.17.12 includes a critical security update addressing multiple vulnerabilities across components such as Foreman, GitPython, Dynaconf, Golang libraries, and others. These vulnerabilities include command injection, remote code execution, information disclosure, denial of service, and authorization bypass issues. The update fixes issues like command injection in job invocations, unauthenticated information disclosure, server-side template injection, and arbitrary code execution via improper validation in GitPython. The advisory also includes bug fixes unrelated to security.
Multiple vulnerabilities have been identified in Foreman, an open-source tool for server lifecycle automation including provisioning, configuration management, and monitoring. Notably, a critical security fix addresses a safemode bypass leading to remote code execution (CVE-2026-96658). Additional vulnerabilities affecting related components such as GitPython and Dynaconf are also addressed in the same advisory. Red Hat Satellite 6.17 for RHEL 9 has an update available that includes these fixes.
A vulnerability in Python's SSLContext.wrap_bio() function causes hostname verification to be skipped silently when hostname checking is enabled but no target server hostname is provided. This flaw allows a man-in-the-middle attacker to spoof trusted endpoints and intercept or tamper with encrypted traffic. The issue affects Red Hat Hardened Images using python3.12 packages. The vulnerability is rated high severity by Red Hat due to its impact on confidentiality and integrity of TLS sessions. A security update is available addressing CVE-2026-19553.
Fortra BoKS Server Agent contains a predictable password generation vulnerability in the adjoin utility. Machine-account passwords generated during Active Directory join or password renewal operations may have significantly less entropy than intended, making them more susceptible to prediction by an attacker who can estimate when the password was generated.
boks_ksllogsd accepts a checksum algorithm name in the MD field of an authenticated KSL start message. Affected releases verify that OpenSSL recognizes the digest name but do not verify that the value fits in a fixed 16-byte checksum context field before copying it. An authenticated KSL client can supply an oversized, OpenSSL-recognized digest name and write beyond the end of the heap allocation.
Langflow-ai langflow version 1.9.3 and earlier contains a critical code injection vulnerability. An authenticated attacker can submit arbitrary Python code to the /api/v1/validate/code endpoint, which is executed on the server without sandboxing or security controls. This allows remote code execution with full impact on confidentiality, integrity, and availability.
A remote, unauthenticated TLS client can make a server crash or call
through a freed pointer if its sni_callback assigns a different context to
SSLSocket.context (the documented way to select a certificate per server
name) and nothing else keeps the original ssl.SSLContext alive. Typical
cases are servers that create an SSLContext per connection or replace it
while connections are open; servers that wrap their listening socket with
it are not affected.
Mitigation: keep a reference to every SSLContext that sets sni_callback for
the lifetime of the server. TLS clients are not affected.
The Python package beautifytext version 1.0.3 (and also 1.0.4 and 1.0.7) contains a malicious backdoor. Upon import or installation, it spawns a detached background process that registers the host with a remote command and control (C2) server and polls it for commands. This backdoor can execute arbitrary shell commands, read and write files on the infected host, and exfiltrate results to the attacker-controlled server. The malicious behavior activates both on import and during package installation, making it stealthy and persistent.
Claude Code selected an API key stored by Claude Code, for example from an earlier `/login` or written directly to its configuration, ahead of the user's valid Claude Enterprise or Team sign-in when fetching the organization's server-managed settings, even though the session itself authenticated with the Enterprise or Team account. When the settings endpoint rejected that stored key, the session started without the organization's server-managed policy (such as permission deny rules, model restrictions and managed-only locks) or, if a previously cached copy existed on the machine, kept applying that stale copy without receiving later policy changes — while continuing to operate as the organization's account. Triggering this required local access to a device with such a stored API key; the no-policy case additionally required that no managed settings had previously been cached. Endpoint-managed (MDM or file-based) settings were not affected. Claude for Enterprise organizations were affected from version 2.0.68; Claude for Work (Team) organizations from version 2.1.38, when server-managed settings became available to them.
Users on standard Claude Code auto-update have received this fix already. Users performing manual updates are advised to update to version 2.1.260 or later.
Thank you to Tamas Voros / NVIDIA AI Red Team for reporting this issue.