Heap-based buffer overflow in open62541 1.5.5 allows unauthenticated remote attackers to crash an OPC UA server by sending a crafted HistoryRead request when the default memory backend is in use, resulting in denial of service. The CVSS vector confirms network-accessible, unauthenticated exploitation with no interaction required, and SSVC classifies it as automatable with a proof-of-concept available. Despite these concerning attributes, EPSS at 0.14% (4th percentile) and absence from CISA KEV indicate limited real-world exploitation activity at time of analysis.
Session hijacking is possible in HUMANIST Digital Human Resources (versions 26.0 through before 26.1) due to sensitive session tokens being transmitted as GET request query string parameters. Exposure occurs through server access logs, browser history, and HTTP Referer headers leaking to downstream systems. No public exploit code has been identified at time of analysis, and no CISA KEV listing exists, but the CVSS integrity and confidentiality impact ratings are both High, reflecting full session takeover potential.
Transient denial of service in Qualcomm Snapdragon's Wi-Fi 6 stack allows adjacent, unauthenticated attackers to crash affected devices by transmitting a malformed Target Wake Time (TWT) channel usage response frame with an undersized packet payload. The CVSS scope change (S:C) indicates the crash propagates beyond the Wi-Fi subsystem to impact the broader host environment. No public exploit code and no CISA KEV listing have been identified at time of analysis.
Command injection in the Zyxel WAX650S Wi-Fi 6 access point firmware enables an authenticated administrator to execute arbitrary OS commands via the "export-cgi" CGI program. All firmware versions through 7.10(ABRM.4)C0 are affected, with full confidentiality, integrity, and availability impact on the device. No public exploit has been identified at time of analysis, and exploitation is gated behind administrator-level credentials, constraining real-world risk to insider threat or post-credential-compromise scenarios.
GeoVision GV-ASManager's unsafe DLL search path resolution allows a local low-privileged attacker who can write to a directory searched by the application to preempt a legitimate library with a crafted DLL, executing arbitrary code under the GV-ASManager process's security privileges when the application is subsequently launched. The root cause is CWE-428 (Unquoted Search Path or Element), where Windows DLL resolution traverses attacker-controllable directories before reaching the intended library location. No public exploit code has been identified and the vulnerability is not listed in CISA KEV, but the full C:H/I:H/A:H CVSS impact reflects total process compromise upon successful exploitation.
Heap-buffer-overflow vulnerabilities in GIMP's APNG file loader and DDS plug-in allow code execution when a user opens a specially crafted image file. The APNG flaw is triggered when the fcTL chunk width exceeds the IHDR width, causing pixel data to overwrite memory beyond the heap allocation; the DDS flaw stems from a bits-per-pixel mismatch in load_layer(). Both affect GIMP as packaged on Red Hat Enterprise Linux 6 through 9. No public exploit identified at time of analysis, and the vulnerability is not listed in CISA KEV.
Path traversal in Zyxel's ZLD-based firewall firmware (ATP, USG FLEX, USG FLEX 50(W), and USG20(W)-VPN series) allows an authenticated administrator to supply a crafted configuration file path to the CLI execution command, bypassing directory restrictions and achieving arbitrary code execution on the device. All four product lines share the same vulnerable CLI mechanism across wide firmware version ranges, substantially broadening the attack surface for organizations running unpatched appliances. No public exploit or CISA KEV listing has been identified at time of analysis, limiting immediate real-world risk to scenarios involving credential compromise or insider threat.
Denial of service in SnailJob 1.7.0 allows authenticated attackers to crash the JVM by exploiting the FuryUtil.deserialize helper's failure to validate the frame_content_size field in Zstandard-compressed frame headers before allocating memory. By storing a base64-encoded crafted payload with an arbitrarily inflated decompressed-size declaration in a retry task argument, an attacker causes an unbounded array allocation attempt when the task traverses the retry-task pipeline, resulting in an unrecoverable java.lang.OutOfMemoryError and full server crash. No public exploit code has been identified at time of analysis, and this vulnerability is not listed in the CISA KEV catalog; a vendor patch has been released as vsj2.0.0.
Path traversal in CVAT versions 1.6.0 through 2.64.0 enables an authenticated, low-privileged attacker with write access to an integrated cloud storage - or permission to add new cloud storage integrations - to overwrite arbitrary files on the server filesystem by embedding traversal sequences in cloud storage file paths processed by the application. The CVSS 4.0 vector (AV:N/AC:L/AT:N/PR:L/UI:N/VI:H) confirms network-accessible, low-complexity exploitation requiring only low-privilege access with high integrity impact. No public exploit or CISA KEV listing has been identified at time of analysis.
Missing authorization middleware on the PUT /api/v1/executions/:id endpoint in Flowise prior to 3.1.3 allows any authenticated workspace member, regardless of assigned RBAC role, to overwrite execution state, data, and metadata for any workflow execution in their workspace. The flaw is a straightforward RBAC enforcement gap - the checkAnyPermission() middleware present on all sibling GET and DELETE routes was omitted from the PUT handler - making privilege escalation trivially reproducible with a single authenticated API call. No confirmed active exploitation (not in CISA KEV), but a working proof-of-concept curl command is published in the vendor's GitHub security advisory, and a vendor-released patch is available in version 3.1.3.
Unauthorized workspace variable disclosure in Flowise (<= 3.1.2) allows any authenticated user or API key - even one explicitly denied the `variables:view` RBAC permission - to exfiltrate all workspace secrets by calling `/api/v1/node-custom-function`. The `$vars` object, populated with static variable values and runtime secrets resolved from `process.env`, is unconditionally injected into every custom JavaScript execution sandbox (utils.ts:1782), while the official `/api/v1/variables` endpoint correctly enforces the permission gate - creating a direct privilege boundary bypass. Depending on workspace configuration, exposed data can include database passwords, JWT signing secrets, SMTP credentials, and cloud provider API keys. No public exploit code has been identified at time of analysis, and this CVE is not listed in CISA KEV.
Improper authorization in ERPNext's CRM module allows authenticated low-privileged users to retrieve CRM opportunity data beyond their permitted scope via the whitelisted API method erpnext.crm.doctype.prospect.prospect.get_opportunities. Versions before 15.115.0 and before 16.26.0 are affected, with patches now available from Frappe. No public exploit code or CISA KEV listing has been identified at time of analysis, though the low-privilege, network-accessible nature of the flaw makes it a realistic insider threat or credential-abuse vector.
Authenticated low-privilege users in Zammad 7.1.0 can access confidential ticket article attachments beyond their authorization scope via a missing authorization check in the attachment cloning endpoint. The flaw, classified as CWE-862 (Missing Authorization), allows any authenticated user to clone and retrieve attachments from tickets they should not have access to, enabling cross-tenant or cross-ticket data leakage in helpdesk environments. No public exploit code or CISA KEV listing has been identified at time of analysis; a patched release (7.1.2) is available from the vendor.
Buffer overflow in the TP-Link Tapo P110 v1 smart plug's HTTP request handling allows an adjacent-network attacker to crash the web service process, resulting in a denial-of-service condition. The device's web service performs a memory copy of HTTP request body data without validating input length against the destination buffer, enabling an oversized payload to corrupt adjacent memory regions. No public exploit code or CISA KEV listing exists at time of analysis; a vendor-released firmware patch is available via the TP-Link download portal.
Path traversal in NVIDIA Triton Inference Server's MLflow plugin allows a low-privileged local user to read, write, or modify files outside the designated model repository by embedding traversal sequences in a model name parameter. All Linux deployments of Triton Inference Server through version 26.02 are affected. No active exploitation has been confirmed (not listed in CISA KEV), EPSS is 0.16% (6th percentile), and SSVC rates exploitation status as none - indicating this is a real but currently low-observed-risk vulnerability that warrants patching on a standard cadence rather than emergency response.
Out-of-bounds read in SSSD's PAM responder allows a local low-privileged attacker to crash the SSSD process and disclose sensitive memory contents by sending a crafted protocol v1 authentication request. Affected systems span all major Red Hat Enterprise Linux releases (6 through 10) and OpenShift Container Platform 4, representing a broad footprint in enterprise Linux environments. No active exploitation is confirmed (not listed in CISA KEV), and the EPSS score of 0.14% at the 3rd percentile reflects a very low observed exploitation probability at time of analysis.
pardus-image-writer before 1.0.4, developed by TÜBİTAK BİLGEM Software Technologies Research Institute for Turkey's national Pardus Linux distribution, permits external control of file name or path (CWE-73), enabling a locally authenticated attacker to redirect file operations outside their intended scope and remove important client functionality. The CVSS vector (AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H, score 7.1) confirms that standard user privileges and no interaction from other users are required, with high integrity and availability impact but no confidentiality impact. No public exploit code or active exploitation has been identified at time of analysis; this is a Turkish CERT (USOM)-reported vulnerability with a confirmed fix at version 1.0.4.
Configuration injection in marimo before 0.23.15 enables a malicious notebook author to silently exfiltrate an operator's OpenAI API key by embedding an attacker-controlled `base_url` in PEP-723 inline script metadata, which the `sanitize_pyproject_dict` function merges into session configuration at higher precedence than the operator's own settings. When the operator opens the crafted notebook and invokes any AI feature - code completion, AI chat, or similar - marimo resolves the injected endpoint while supplying the operator's `OPENAI_API_KEY` as the Authorization credential, transmitting it to the attacker's server. Critically, no cell execution is required; the exfiltration fires on the AI request itself. No public exploit confirmed at time of analysis, though VulnCheck's detailed advisory materially lowers the reproduction barrier. Patch is available in marimo 0.23.15.
SQL injection in Atals-Livre exposes its database to manipulation by authenticated high-privileged users through an unsanitized GET parameter in the supp() deletion helper function. Exploitation enables unauthorized SQL operations including arbitrary record deletion and data extraction against any database instance backing the application. A public Gist writeup exists that appears to document reproduction steps; no CISA KEV listing confirms active exploitation at time of analysis.
Improper TLS certificate validation in multiple Lenovo XClarity Orchestrator (LXCO) 2.2.0 microservices exposes sensitive management communications to interception by an adjacent network attacker capable of positioning themselves for a machine-in-the-middle attack. The flaw (CWE-295) allows the attacker to bypass HTTPS trust enforcement under unspecified triggering circumstances, potentially capturing credentials, tokens, or orchestration data exchanged between LXCO microservices. No public exploit code or CISA KEV listing exists at time of analysis; Lenovo has released a patch via advisory HT509976.