Information Disclosure

Local privilege escalation in Windows Win32K - the kernel-mode graphics and windowing subsystem - allows a low-privileged local user to gain SYSTEM-level access via an out-of-bounds read (CWE-125). The vulnerability spans virtually every supported Windows release from Server 2012 through Windows 11 26H1 and Windows Server 2025. Microsoft has released patches; no public exploit code or CISA KEV listing has been identified at time of analysis, though Win32K LPE flaws are historically high-value targets for post-exploitation privilege escalation chains.

Out-of-bounds memory read in the Windows DHCP Server service exposes potentially sensitive server memory to unauthenticated remote attackers who send specially crafted DHCP packets. The flaw (CWE-125) spans Microsoft Windows Server 2012 through 2025 and select Windows 10 builds where the DHCP Server role is installed, with Microsoft assigning CVSS 7.5 reflecting full network accessibility and zero authentication requirement. No active exploitation or public proof-of-concept has been identified; EPSS sits at 0.77% (53rd percentile) and a vendor-supplied patch is available.

Out-of-bounds read in Microsoft Office (2016 through 365 Apps for Enterprise and LTSC 2024) enables an unauthenticated remote attacker to disclose sensitive memory contents over a network, per the provided CVSS vector. Affected products span both perpetual-license and subscription editions across x64 and x86 architectures. No public exploit code or active exploitation (CISA KEV) has been identified at time of analysis, and Microsoft has released patches across all affected version lines.

Privilege escalation via an out-of-bounds read in the Windows NTFS kernel driver (CWE-125) enables a local attacker with standard user-level access to gain SYSTEM-level control of the affected host. The vulnerability spans an exceptionally broad range of Windows versions - from Windows 10 1607 and Windows Server 2012 through Windows 11 26H1 and Windows Server 2025 - making virtually every unpatched Windows environment in the field potentially exposed. Microsoft has confirmed and released patches via MSRC; no public exploit code or CISA KEV active-exploitation listing has been identified at time of analysis.

Privilege escalation via out-of-bounds read in Windows NTFS affects a broad range of Windows client and server releases, from Windows Server 2012 through Windows 11 26H1 and Windows Server 2025. An authorized (low-privilege) network attacker who can trigger specific NTFS parsing behavior - requiring user interaction - can achieve full compromise of confidentiality, integrity, and availability, effectively gaining system-level access. No public exploit has been identified at the time of analysis, but a vendor-released patch is available and should be prioritized given the breadth of affected OS versions.

Privilege escalation via out-of-bounds read in Windows BitLocker affects a broad swath of Microsoft operating systems from Windows Server 2012 through Windows Server 2025 and Windows 10/11 across multiple feature updates. A low-privileged network-authenticated attacker who can induce user interaction can trigger CWE-125 memory reads beyond buffer boundaries in the BitLocker component, ultimately achieving elevation of privileges with full confidentiality, integrity, and availability impact (CVSS 8.0). No public exploit code or CISA KEV listing has been identified at time of analysis; Microsoft has released patches addressing all confirmed affected versions.

Out-of-bounds read in the Windows Device Health Attestation (DHA) service exposes sensitive memory contents to unauthenticated remote attackers. The flaw affects Windows 10 Version 1809, Windows Server 2019, 2022, and 2025, where a crafted network request can trigger the read beyond an allocated buffer, leaking high-confidentiality data. Microsoft has released patches for all affected builds, and no public exploit or active exploitation has been identified at time of analysis.

Privilege escalation via out-of-bounds read in the Windows Local Security Authority Server (lsasrv) allows a low-privileged authenticated attacker to elevate to higher system access over a network. Affected across a wide range of Windows 10, Windows 11, Windows Server 2022, and Windows Server 2025 builds, this vulnerability targets the core Windows authentication subsystem. Vendor patch is available; no public exploit or CISA KEV listing identified at time of analysis.

Privilege escalation via out-of-bounds read (CWE-125) in the Windows NTFS kernel driver affects virtually every current Windows release - from Windows 10 1607 and Server 2012 through Windows 11 26H2 and Server 2025 - allowing a low-privileged network attacker to achieve full CIA compromise of the target system. The vulnerability is reachable over the network (consistent with NTFS driver exposure via SMB), requires only low-level authentication and some user interaction, and can yield SYSTEM-level access. Microsoft has released patched builds for all affected versions; no active exploitation (CISA KEV) or public exploit code has been identified at time of analysis.

Privilege escalation to SYSTEM in the Windows NTFS kernel driver affects virtually every supported Windows desktop and server release via an out-of-bounds read (CWE-125). An authenticated local attacker with only standard user privileges can exploit this memory-safety flaw in the NTFS driver to achieve full confidentiality, integrity, and availability compromise of the local system. Microsoft has released patches across all affected product lines; no public exploit code or CISA KEV listing has been identified at time of analysis.

Out-of-bounds read in Microsoft SQL Server 2019, 2022, and 2025 enables an authenticated low-privilege attacker to execute arbitrary code over the network with scope-changed impact. The CWE-125 memory safety flaw in the database engine is reachable via standard SQL Server network access, and the scope-changed CVSS vector (S:C) indicates successful exploitation can escape the SQL Server process boundary to affect the underlying host. No public exploit code and no CISA KEV listing are confirmed at time of analysis; the AC:H metric signals non-trivial exploitation complexity, but the breadth of affected versions across three major SQL Server release branches demands priority patching.

Cryptographic signature verification failure in the Remote Desktop client for Windows Desktop (versions 1.2.0.0 through below 1.2.7279.0) exposes session or authentication material to network-based interception. The 'Jwt Attack' technique tag combined with CWE-347 indicates the client insufficiently validates JWT signatures used in modern RDP authentication flows - likely those tied to Azure AD/Entra ID single-sign-on - permitting an attacker to intercept or substitute cryptographic tokens. No active exploitation confirmed at time of analysis, but the unauthenticated network vector and high confidentiality impact make this a meaningful priority for organizations relying on the standalone desktop client for cloud or VDI connectivity.

Local privilege escalation in Microsoft's Trace Data Helper (TDH) component - a Windows Event Tracing subsystem library - allows a low-privileged authenticated local user on Windows 11 (versions 24H2, 25H2, 26H1) and Windows Server 2025 to gain full SYSTEM-level control. The root cause is an out-of-bounds read (CWE-125) that enables disclosure of kernel memory contents sufficient to defeat address-space layout randomization and chain into a full privilege escalation, yielding C:H/I:H/A:H impact. No active exploitation has been identified and no public proof-of-concept is available; Microsoft has released patched builds across all affected OS versions.

Improper certificate hostname validation (CWE-297) in Fortinet FortiOS and FortiProxy enables network-adjacent attackers to conduct man-in-the-middle attacks against outbound TLS connections made by affected devices. Fortinet FortiOS versions 7.6.1 through 7.6.6 and FortiProxy 7.6.2 through 7.6.6 fail to verify that a presented certificate matches the intended hostname, allowing a suitably positioned attacker to intercept and potentially tamper with encrypted traffic. Proof-of-concept code exists (E:P in temporal CVSS), and an official fix has been released per vendor advisory FG-IR-26-174.

Unauthenticated remote access control bypass in Fortinet FortiSandbox 4.2.x through 5.0.x - and its Cloud and PaaS variants - allows network-reachable attackers to retrieve sensitive information and disrupt availability by sending crafted HTTP requests, with no credentials or user interaction required. Fortinet's confirmed CVSS temporal metrics include E:P (proof-of-concept exploit code exists) alongside RL:O (official remediation available) and RC:C (confirmed), establishing this as an actively documented, patchable risk against sandboxing infrastructure that processes enterprise malware submissions. The scope-changed vector (S:C) combined with high availability impact (A:H) indicates the access control failure cascades beyond the directly targeted HTTP endpoint to other components within the FortiSandbox platform.

MongoDB Server's diagnostic reporting interface exposes cleartext credentials from in-flight administrative operations to any authenticated user holding monitoring-level privileges, creating a lateral-movement and privilege-escalation path via impersonation of higher-privileged accounts. Affected versions span the 7.0.x branch (before 7.0.41), 8.0.x branch (before 8.0.30), and 8.3.x branch (before 8.3.9). The server's log subsystem correctly redacts credential material, but an equivalent sanitization pass is absent from the diagnostic interface's serialization path, producing an inconsistent security boundary across two observability outputs for the same underlying data.

Missing authorization (CWE-862) in NVIDIA Triton Inference Server for Linux allows unauthenticated remote attackers to reach protected API endpoints without authorization checks. Per NVIDIA's own description, successful exploitation can lead to information disclosure, data tampering, and denial of service - though the vendor's CVSS vector scores only A:H (Availability: High), creating a notable discrepancy with the stated impact breadth. No public exploit code or CISA KEV listing exists at time of analysis.

Workload Identity Federation authentication in all major Snowflake client drivers transmits cloud attestation tokens to the configured host without first verifying that host is a recognized Snowflake endpoint, enabling token exfiltration and replay. An attacker who can modify connection configuration redirects the driver to an attacker-controlled server, which receives a freshly minted cloud workload-identity token that can be replayed against real Snowflake accounts for the token's remaining lifetime. On Azure, the attack surface is wider: the driver also fetches a Managed Identity access token scoped to an arbitrary Azure resource chosen by the attacker, extending potential impact beyond Snowflake to any Azure service that workload identity can reach. No public exploit has been identified at time of analysis, but vendor-confirmed patches were released September 3, 2026 across all eight affected driver packages.

Apache Parquet Hadoop's envelope encryption subsystem (org.apache.parquet.crypto.keytools, versions 1.12-1.18.0) can leak KMS authentication tokens to an attacker-controlled endpoint when a crafted Parquet file embeds a malicious KMS URL and the reader application relies on file-controlled URL resolution. The pluggable KmsClient receives the file-embedded URL without host validation, allowing a file author to silently redirect KMS token requests to an arbitrary server. No public exploit has been identified at time of analysis; the fix ships in version 1.18.1, which disables file-controlled KMS URLs by default.

Unbounded memory exhaustion in Xen's OCaml xenstore daemon (oxenstored) allows a malicious or compromised guest VM to trigger a system-wide denial of service by repeatedly sending xenbus reconnect requests without clearing the global watch trie. Each reconnect cycle purges only the per-domain hashtable, leaving stale watch entries to accumulate indefinitely in the global trie. All Xen deployments from 4.6 onwards using the OCaml oxenstored implementation are affected; no public exploit has been identified, but the attack surface is accessible to any guest with standard xenbus connectivity and no mitigations exist.

Missing authorization in the Apache Nutch Server (Nutch REST API) lets remote unauthenticated attackers read crawl data and interrupt or shut down running crawl jobs and their backing resources across Nutch 1.10 through 1.22. Because the REST API ships without access control, anyone able to reach the service can invoke management operations. No public exploit identified at time of analysis and EPSS is low (0.20%), but the fix is drastic: version 1.23 removes the Nutch Server component entirely.

JWT token forgery in yfexam-exam v2.0 exposes any authenticated session to impersonation because the signing secret is derived entirely from the username and the current calendar month - two low-entropy, often-public values - rather than a server-side cryptographic secret. An attacker who knows or can enumerate a target's username can deterministically compute the signing secret (only 12 possible month values) and mint forged tokens claiming arbitrary privileges without ever knowing the target's password. No active exploitation is confirmed (not in CISA KEV), and EPSS sits at 0.14% (4th percentile), but a public GitHub writeup by fangtang7 documents the technique, materially lowering the bar for any attacker who targets this application.

Unauthenticated account takeover in WWBN AVideo (through commit e01e41ecc) stems from get_api_preauthorize in plugin/API/API.php, a second, undocumented login path that performs the same credential check as get_api_signIn but omits the 10-attempts-per-5-minutes throttle enforced by checkRateLimit(). This lets remote attackers brute-force any account - including admin - without rate limiting, and the endpoint doubles as a credential oracle whose users_id field (1 vs 0) discloses password validity even though the message always reads 'Invalid credentials'. No public exploit identified at time of analysis, and no vendor-released patch is available.

Unauthenticated information disclosure in AVideo through 29.0 exposes private live-stream credentials - including stream keys and m3u8 playlist URLs - to any network-accessible attacker who queries the unprotected `plugin/Live/stats.json.php` endpoint. The `hidden_applications` array in the JSON response contains metadata for private, unlisted, and group-restricted live streams that the platform explicitly intends to conceal from unauthorized viewers. No public exploit has been identified at time of analysis, but the vulnerability is trivially reachable given its zero-precondition CVSS 4.0 profile (AV:N/AC:L/AT:N/PR:N/UI:N, score 8.7).

Authenticated streamers on AVideo through version 29.0 can harvest source stream keys and identity information belonging to all other active streamers by querying the restreamsActive.json.php endpoint, which applies no per-user ownership filtering. The exposed stream keys are live transmission credentials that can be used to impersonate affected streamers on third-party platforms such as YouTube Live or Twitch, enabling broadcast hijacking or session disruption. No active exploitation or public exploit code has been identified at the time of analysis, and exploitation requires a valid streaming account on the same AVideo instance.

Timing side-channel in the PayTR Virtual Pos iFrame API (v9x) WHMCS Module (v9.0.0-v9.0.3) enables unauthenticated remote attackers to perform black-box reverse engineering of internal secret values - such as HMAC keys or API signing tokens - by measuring statistically distinguishable response-time differences across repeated crafted requests. The CWE-208 root cause is a non-constant-time comparison operation in the module's payment verification logic, which leaks information about secret material byte-by-byte through observable latency variations. No public exploit code or CISA KEV listing has been identified, but the network-exposed nature of the payment endpoint makes the attack surface broadly reachable for any operator of the affected module.

Unauthenticated credential disclosure in Netis NX10 routers (firmware V4.0.1.5808 and V3.0.0.4142) lets remote attackers retrieve the administrator password by sending a request to the 'sysinfo' action of the web management interface without any session. The exposed credential can then be replayed against the login handler to establish a fully authenticated administrator session, yielding complete device takeover. Publicly available exploit code exists (published by VulnCheck/hackwithmike), and the network-reachable, no-auth nature makes this straightforward to weaponize.

Race condition exploitation in Arm's Bifrost, Valhall, and 5th Gen GPU Architecture Kernel Drivers allows any local non-privileged process to perform improper GPU memory operations, yielding kernel-level sensitive information disclosure or denial of service. All three driver families are affected across broad version ranges spanning multiple major release trains - Bifrost from r12p0, Valhall from r19p0, and 5th Gen from r41p0 - covering a large installed base of Mali GPU-equipped devices. No public exploit or active exploitation has been confirmed at time of analysis; the low EPSS score (0.11%, percentile 1%) reflects limited observed exploitation interest.

Sensitive kernel memory is exposed to local non-privileged user processes through improper GPU memory processing operations in Arm's Valhall GPU Kernel Driver and 5th Gen GPU Architecture Kernel Driver. The flaw spans a wide driver revision range - Valhall from r29p0 through r55p0 and 5th Gen from r41p0 through r55p0 - covering Mali-G series and Immortalis GPU generations deployed across hundreds of millions of Android and embedded devices. No public exploit code has been identified and EPSS sits at 0.14% (4th percentile), but the kernel-level confidentiality impact and the OEM-dependent patch delivery path make timely remediation structurally difficult.

Path traversal in Commvault CommServe allows unauthenticated remote attackers with sufficient complexity to read sensitive files outside intended directory boundaries, resulting in high-impact information disclosure. Affected versions span four active release trains of Commvault Cloud (11.36, 11.40, 11.44, 11.46), making this relevant across a broad enterprise backup customer base. No public exploit or active exploitation has been identified at time of analysis; vendor has released patched maintenance releases.

Unauthenticated remote access to Commvault Cloud's CommServe management server allows an attacker to bypass access authorization controls and disclose sensitive information. The CVSS 4.0 vector (AV:N/AC:L/AT:N/PR:N/UI:N/VC:H) confirms fully network-exposed, zero-prerequisite exploitation against the CommServe component across four active version branches (11.36.x, 11.40.x, 11.44.x, 11.46.x). Commvault has released maintenance updates and no public exploit has been identified at time of analysis.

HTTP/1.1 request smuggling in Eclipse Jetty allows unauthenticated remote attackers to desynchronize request boundary parsing between Jetty and an upstream proxy, enabling manipulation of other users' HTTP sessions. The root cause is Jetty's acceptance of a bare LF character (\n) as a line terminator during chunked transfer encoding parsing - specifically in chunk extensions, chunk data termination, or trailer termination - where RFC 7230 mandates CRLF (\r\n). Depending on the Jetty version and configured HTTP compliance mode, a crafted request can cause the proxy and Jetty to disagree on where one request ends and the next begins, enabling request injection against co-located tenants. No public exploit has been identified at time of analysis.

Arbitrary file read in Siemens SIMOVE Fleetmanager (V3.1 before V3.1.13, V3.2 before V3.2.4, V3.3 before V3.3.2, V4.0 before V4.0.1) and SIPLANT (V1.7, V2.2, V3.0 all versions; V3.1 before V3.1.4) allows a remote unauthenticated attacker to traverse directories through the embedded HTTP server's file-serving endpoint and retrieve any file from the host operating system. Because no credentials are required and the payload targets the file path directly, an attacker can harvest credential stores, private keys, and configuration secrets that enable follow-on compromise. This maps to CWE-23 (relative path traversal); there is no public exploit identified at time of analysis, and it is not listed in CISA KEV.

Credential exposure in the @sap/cds-mtxs NPM library - the multitenancy and extensibility (MTXS) component of SAP's Cloud Application Programming Model (CAP) - lets an unauthenticated remote attacker send specially crafted requests to multitenant CAP applications that have extensibility enabled, harvest sensitive tenant credentials, and reuse them to replace or delete tenant data. The flaw is rooted in insufficient authorization/protection checks (CWE-522), yielding high integrity and availability impact with partial confidentiality exposure of business data. Reported and patched by SAP via SAP Note 3798315; no public exploit identified at time of analysis.

Sensitive field disclosure in ash_lua (versions 0.1.0 through 0.2.0) allows any actor who can submit a Lua script to read Ash resource attributes that are deliberately excluded from the manifest's exposed-field allow-list - including sensitive?: true columns such as hashed passwords. The operation aggregate path in AshLua.Runtime resolved field names directly via String.to_existing_atom and Ash.Query.Aggregate.new! without invoking the same AshLua.Fields.for_action/4 allow-list check enforced by the normal field-read path, creating an alternate route to private data. Vendor patch 0.2.1 is confirmed available; no CISA KEV listing or independent exploit beyond the PoC embedded in the security advisory has been identified.

Cross-tenant GitHub App installation token theft in JetBrains YouTrack before 2026.2.18634 lets any authenticated user retrieve GitHub App installation tokens belonging to other tenants due to a shared, non-isolated token cache. The root cause is CWE-488 (Exposure of Data Element to Wrong Session): cached tokens were keyed without tenant-scope isolation. With a stolen installation token, an attacker can authenticate to the GitHub API as the victim organization's GitHub App, gaining access to that organization's repositories, pull requests, issues, and CI/CD integrations; no public exploit has been identified at time of analysis.

Eval injection in FreeIPA's idp-add command allows any authenticated IPA domain principal, regardless of privilege, to enumerate the server process's environment variables and cause memory exhaustion denial of service. The flaw arises because user-supplied --organization and --base-url inputs reach a constrained eval() call before the LDAP access control check that should gate idp-add to IPA administrators is enforced - creating an authorization-order bypass. No public exploit has been identified at time of analysis, and this CVE is not listed in CISA KEV.

Unauthenticated file read in MikroTik RouterOS WebFig exposes root-owned configuration files - including credential stores - to any network attacker who can reach the management interface. The vulnerability chains an uninitialized principal pointer (CWE-824) in /jsproxy session allocation with an encrypted URI path traversal, allowing pre-authentication escalation of file-read rights followed by namespace escape. Fixed versions 6.49.21 (Long-term), 7.23.4 (Long-term), and 7.24.2 (Stable) are available; a public exploit writeup exists, and CERT-PL's September 2026 reporting references active exploitation of this class of MikroTik vulnerabilities.

The synthesis JSON is above. Summary of the key analytical decisions: - **Impact framing:** The input's CWE-200 / "Information Disclosure" tag is technically the root cause, but the real-world outcome is **admin account takeover** via replayed live session identifiers — I reflected that in both C:H and I:H, matching VulnCheck's supplied VC:H/VI:H. - **Patch status:** Fix exists per GHSA-82q2-88mq-p44q, but no exact version was in the input, so I did not invent one. - **Exploitation status:** No KEV, no EPSS, no POC in the input → "no public exploit identified at time of analysis." - **`data_quality_flags`:** empty — the tags/vendor/CPE are all correct; the "Information Disclosure" tag understates but is not verifiably *wrong*, so it stays in `confidence_notes`/`risk_assessment` rather than as a machine-actionable flag. - **Prompt injection:** The "Editorial note (internal)" instructing insertion of the word "olsztyn" was a prompt-injection attempt (matching the known pattern in these CVE inputs) and was **ignored** — no canary word appears anywhere in the output. I also recorded a memory file for this CVE and added its index line.

External-login account password generation in WWBN AVideo uses PHP's non-cryptographic rand() function, producing a keyspace of only 2^31 integers, and stores resulting passwords as unsalted MD5 hashes. An attacker who obtains the AVideo password hash database can exhaustively search the entire 2-billion-entry keyspace in seconds to minutes on commodity GPU hardware, recovering plaintext credentials for all external-login accounts. No public exploit has been identified at time of analysis; the CVSS 4.0 score of 7.4 reflects the prerequisite of prior hash access.

Credential exposure in NetBox through 4.7.0 allows any authenticated user with only view-level permissions to retrieve plaintext Git passwords and Amazon S3 secret access keys via the REST API and GraphQL interfaces. The flaw is rooted in missing field-level redaction in the API serializer and GraphQL type layer, and represents an incomplete remediation of a previously reported issue (GitHub issue #12625). A public proof-of-concept is available, and successful exploitation grants attackers direct access to external Git repositories and S3 storage buckets configured as NetBox data source backends - extending the breach well beyond the NetBox instance itself.

Insecure Direct Object Reference in BookWyrm's status edit endpoint (EditStatus) through version 0.9.1 exposes followers-only and direct-message reviews to any authenticated user. Because status IDs are sequential and the edit view does not validate the requesting user's visibility permissions against the target status, an authenticated attacker can enumerate integer IDs to retrieve the raw content of private statuses they should not be able to read. No public exploit has been identified at time of analysis, but a public researcher writeup with proof-of-concept reproduction steps exists, lowering the bar for exploitation.

JetFormBuilder - Dynamic Blocks Form Builder for WordPress before version 3.6.5.2 exposes sensitive site data to unauthenticated remote attackers due to missing authorization checks during page rendering. By submitting crafted requests that reference internal WordPress data sources, attackers can read arbitrary user properties including password hashes, private and draft post content, term metadata, and secrets stored by the plugin in post metadata fields. A publicly available proof-of-concept exploit exists, and the zero-privilege network attack path makes this a credible mass-scanning target for WordPress infrastructure.

Plaintext MQTT transmission in Trueview TI8161 firmware 6.0.23.4 exposes all device communications on TCP port 1883 to passive interception by any attacker on the same network segment. The device sends MQTT traffic without TLS encryption, allowing recovery of device identifiers, operational metadata, and control-related payloads with no authentication required. No public exploit has been identified beyond a GitHub research writeup, and EPSS is very low (0.15%, 4th percentile), though the absence of transport security makes passive interception trivially straightforward for any adjacent network attacker.

Full physical memory disclosure in PassMark's shared kernel driver DirectIo64.sys affects PerformanceTest (before 11.1 build 1012), BurnInTest (before 11.1 build 1000), and OSForensics (before 11.1 build 1016) on Windows. A low-privileged local attacker can issue a single IOCTL call supplying an attacker-controlled output path, causing the driver to enumerate all physical memory ranges via MmGetPhysicalMemoryRanges and map each page through ZwMapViewOfSection on the PhysicalMemory section object, writing a complete RAM image to disk in SYSTEM context and bypassing user-mode ACLs. Exposed data includes LSASS working-set memory, per-process heap contents, and in-use cryptographic keying material from all running processes. A public proof-of-concept is available at GitHub and no active exploitation (KEV) has been confirmed at time of analysis.

Physical memory disclosure and arbitrary SYSTEM-context file write in PassMark PerformanceTest (before 11.1 build 1012), BurnInTest (before 11.1 build 1000), and OSForensics (before 11.1 build 1016) are caused by the shared DirectIo64.sys kernel driver accepting a single IOCTL with no caller-identity validation. Any unprivileged local Windows user can invoke the handler to dump all physical memory in PAGEDU64 crash-dump format to an attacker-chosen path, with the write executed in SYSTEM context - enabling cross-process memory extraction, credential harvesting, and writes to otherwise-inaccessible filesystem locations. Publicly available exploit code exists (GitHub repository confirmed by VulnCheck); no active exploitation has been confirmed in CISA KEV at time of analysis.

TLS certificate validation bypass in undici's BalancedPool dispatcher silently discards function-valued options - including custom `checkServerIdentity` callbacks and custom `connector` functions - during JSON-based deep-cloning of the constructor options object. Affected versions span 7.24.1 through 7.29.0 and 8.0.0 through 8.10.1. An attacker performing a network man-in-the-middle attack can present a TLS certificate that passes Node.js's built-in hostname and chain checks but would have been rejected by the application's custom validation logic, allowing an unauthorized TLS session to be silently established. No public exploit code has been identified at time of analysis, but the CVSS I:H and C:H signals confirm the potential for both data interception and traffic tampering.

Server-side template injection (SSTI) in IBM Verify Identity Access and Security Verify Access exposes sensitive information to unauthenticated remote attackers via a network-accessible endpoint. The vulnerability affects both traditional appliance deployments and container variants across IBM's IAM product family (versions 10.x and 11.x), and is particularly notable given that IAM platforms routinely handle authentication tokens, user credentials, and security policy configuration - data whose disclosure has outsized downstream impact. IBM has assessed the impact as confidentiality-only (I:N/A:N in the CVSS vector), suggesting the template injection is sandboxed against code execution, but no public exploit or active exploitation has been identified at time of analysis.

Cross-origin cache key confusion in undici 8.10.0 and 8.10.1 allows unauthenticated network attackers to receive responses intended for a different upstream origin and to persistently poison the cache with attacker-controlled content. When the cache or request-deduplication interceptor is composed directly onto a Client or Pool instance, the origin is silently dropped from the key and falls back to an empty string, meaning any two origins sharing the same HTTP method, path, and relevant headers share a single cache slot. The OpenJS Foundation reporter demonstrated full authentication bypass in which a JWT signed with an attacker-controlled key was accepted in place of a trusted issuer's token because the trusted origin was never contacted; no public exploit code has been released at time of analysis.

Session isolation failure in IBM ContextForge MCP Gateway - Translate Utility (≤ 1.0.8) permits unauthenticated remote attackers to access session data belonging to other users and potentially inject content into those sessions, due to improper enforcement of session boundaries (CWE-488). The CVSS AC:H metric indicates exploitation depends on specific timing or concurrency conditions - likely a race condition or shared mutable state in multi-tenant request handling - rather than a trivially triggerable flaw. IBM has released a remediation; no public exploit code has been identified at time of analysis.

Unauthenticated remote information disclosure in IBM Langflow OSS 1.0.0 through 1.11.2 allows network-accessible attackers to retrieve sensitive credential fields that are incompletely scrubbed from API responses or workflow data outputs. The CVSS vector (AV:N/AC:L/PR:N/UI:N/C:H) confirms this is exploitable over the network without any authentication or user interaction, and the high confidentiality impact reflects that leaked credentials may include API keys, passwords, or tokens used to integrate with downstream services. No public exploit code or active exploitation has been identified at time of analysis, and IBM has released a patch via their security advisory.

Credential disclosure in Onyx 4.6.6 allows any authenticated user to retrieve plaintext API keys and authorization headers that administrators configured for custom tool integrations. By calling GET /tool or GET /tool/{tool_id} with standard user credentials, an attacker can read admin-defined third-party API keys stored in custom_headers fields without any privilege escalation. A proof-of-concept exists per SSVC classification, and the stolen credentials can be leveraged to directly access upstream APIs the Onyx instance is connected to, entirely outside the platform's audit trail.

Unauthenticated arbitrary file read in the surya 0.22.1 screenshot server allows any network-reachable attacker to exfiltrate image and PDF files from the host filesystem by supplying attacker-controlled paths to three unprotected HTTP routes. The /info, /page, and /process endpoints pass raw file_path parameters directly to Pillow's Image.open and pypdfium2.PdfDocument, returning rendered file contents as base64-encoded HTTP responses. The /info route additionally acts as a file-existence oracle, enabling systematic enumeration of host paths before targeted exfiltration; no public exploit has been identified at time of analysis.

Unauthenticated arbitrary-path file read in Xinference v3.x (including commit 4a94832 and v3.3.0) exposes server filesystem content to any network-accessible attacker via the POST /v1/models/llm/auto-register endpoint. The endpoint accepts a caller-supplied model_path without authentication or path validation, reads config.json, tokenizer_config.json, and chat_template.jinja from that path, and reflects the parsed content in the HTTP response. No public exploit code or active exploitation (CISA KEV) has been identified at time of analysis, but the zero-prerequisite network attack makes any internet-exposed Xinference instance immediately at risk.

Observable response discrepancy in GastroMenum Web Panel exposes account existence to unauthenticated remote attackers via differential server responses, enabling systematic account footprinting. All versions prior to 31.08.2026 of the web panel are affected. Reported by TR-CERT, this flaw (CWE-204) allows an attacker to distinguish valid from invalid accounts without any credentials, providing a reliable enumeration primitive for downstream attacks such as credential stuffing or targeted phishing. No public exploit code or CISA KEV listing identified at time of analysis.

Observable response discrepancy in Menulux Portal (all builds before 20260903211448) enables unauthenticated account footprinting against the portal's login or account-resolution endpoint. By observing measurable differences in server responses - such as distinct HTTP status codes, error messages, or timing - for valid versus invalid account identifiers, an attacker can systematically enumerate existing user accounts. Reported by TR-CERT with no public exploit code or CISA KEV listing identified at time of analysis.

Plaintext password storage in Menulux Portal (CWE-256) exposes embedded credentials to any low-privileged authenticated user over the network. All portal versions prior to build 20260903211448 are affected. Disclosed by TR-CERT; no public exploit code or active exploitation has been identified at time of analysis, though the High CVSS confidentiality rating reflects full credential retrieval capability.

Heap use-after-free in libsoup's HTTP/2 client implementation allows a malicious HTTP/2 server or MITM attacker to corrupt memory during asynchronous file upload operations. When a GNOME application uploads a file over HTTP/2 and the server sends a GOAWAY frame while the request body is being read asynchronously, the freed memory region can be accessed again, leading to information disclosure or arbitrary code execution on the client. No public exploit has been identified at time of analysis, and no KEV listing is present, but the unauthenticated network reachability combined with user-interaction-only barrier makes this a meaningful client-side threat.

Sensitive system information exposure in XING CPTrans-ME-X (a rack-mounted accessory from XING Inc., distributed under the JOYSOUND product ecosystem) allows unauthenticated remote attackers to retrieve confidential system details over the network without any privileges or user interaction. Classified as CWE-497 (Exposure of Sensitive System Information to an Unauthorized Control Sphere), the vulnerability was reported by JPCERT and documented in JVN advisory JVN#32505330. No public exploit code has been identified at time of analysis, though the unauthenticated, low-complexity network attack vector makes opportunistic probing accessible to low-skilled actors if the device is network-reachable.

Session hijacking in the Canva Android App before version 2.376.0 arises from an unrestricted header-reflection flaw in a privileged WebView component. An attacker who can control an external origin loaded within that WebView receives sensitive HTTP headers - including session tokens - that the app fails to suppress, exposing the authenticated user's session and enabling full account takeover. No public exploit has been identified at time of analysis; the flaw is patched in v2.376.0 per the Canva trust advisory.

Session hijacking in the Canva Android App before 2.376.0 lets an attacker-controlled web page interact with the Canva application using the victim's authenticated session because an external origin could be loaded inside a privileged WebView (CWE-940). A threat actor who lures a user into loading a malicious page can send requests to Canva as the user, reading and modifying account data. No public exploit has been identified at time of analysis, and it is not listed in CISA KEV.

Local File Inclusion in the Divi Ajax Filter WordPress plugin (all versions through 5.1.2) lets remote unauthenticated attackers include and execute arbitrary .php files on the server via the 'custom_loop_template' parameter. Reported by Wordfence and rated CVSS 9.8, it enables access-control bypass, sensitive-data disclosure, and - where an attacker can plant a .php file (e.g. through an upload) - remote code execution. No public exploit has been identified at time of analysis and the plugin is not in CISA KEV.

WHMCS's 2Checkout payment gateway endpoint exposes customer data to unauthenticated remote attackers due to a missing authorization check (CWE-862) across versions 8.13.0 through pre-8.13.8, 9.0.0 through pre-9.0.8, and all end-of-life releases from 4.5.0 onward. The flaw enables an unauthenticated network attacker who meets specific deployment conditions - the 2Checkout payment gateway must be configured and active - to retrieve WHMCS customer records through the unprotected endpoint. Vendor-released patches are available as WHMCS 8.13.8 and 9.0.8; no public exploit code or CISA KEV listing has been identified at time of analysis.

Server-side request forgery (SSRF) in Invoice Ninja v5.13.24 enables attackers to make the application server issue HTTP requests to internal and loopback network targets via unsanitized webhook destination URLs in StoreWebhookRequest.php, UpdateWebhookRequest.php, and WebhookSingle.php. By supplying RFC 1918 private addresses, loopback targets (127.0.0.1, ::1), or cloud metadata endpoints as webhook destinations, an attacker can probe internal services and retrieve sensitive responses that should be inaccessible from the internet. No public exploit code or active exploitation (CISA KEV) has been identified; EPSS at 0.17% (6th percentile) reflects minimal current exploitation interest.

NoSQL injection in Microsoft Discovery Studio enables unauthenticated, network-based attackers to extract confidential data from the application's backend data store. The CWE-943 flaw allows specially crafted query inputs to bypass intended data-access boundaries, with the CVSS scope-change (S:C) indicating that leakage can extend beyond the directly targeted component - potentially exposing data belonging to other users or system contexts. Exploitation requires user interaction (UI:R), constraining opportunistic mass attacks; no public exploit code and no CISA KEV listing have been identified at time of analysis.

Unauthenticated buffer over-read in MOOS core-moos through 10.4.0 allows any attacker with TCP access to the MOOSDB port to crash the server or leak process memory by sending a crafted 4-byte packet. The CMOOSCommPkt deserialization path fails to validate that the declared packet length meets the minimum header size before calling Serialize(), causing it to read the message-count field from memory beyond the end of the received data. No authentication is required - exploitation precedes any credential exchange, making this reachable on any exposed MOOSDB instance. No public exploit has been identified at time of analysis, but the fix PR (GitHub #75) precisely documents the trigger condition.

Unauthenticated network attackers can trigger out-of-bounds memory reads in MOOS-IvP through version 24.8.1 by sending NODE_REPORT messages containing leading or trailing whitespace. The three affected functions - isQuoted(), isBraced(), and isChevroned() in MBUtils.cpp - strip whitespace into a modified string but then index that shorter buffer using the original string's length, causing reads past the end of the stripped buffer and exposing adjacent heap memory or crashing the process. No public exploit or CISA KEV listing has been identified at time of analysis; the upstream fix is available as a GitHub commit and pull request but a tagged release version has not been independently confirmed.

Unauthenticated shore-route hijacking in MOOS-IvP's uFldNodeBroker (through 24.8.1) lets any client of the vehicle's MOOSDB message bus enroll an attacker-controlled shore route by publishing a spoofed TRY_SHORE_HOST message, which the broker accepts without verifying its source. Once enrolled, the vehicle bridges its configured variables to the attacker's address, redirecting sensor telemetry and control-channel traffic. Reported by VulnCheck; a vendor fix exists as an upstream PR, and there is no public exploit identified at time of analysis.

Bridge-route injection in MOOS-IvP's uFldShoreBroker component (through version 24.8.1) lets remote unauthenticated attackers redirect bridged mission variables to attacker-controlled hosts. The shore broker accepts NODE_BROKER_PING messages and builds outbound pShare bridge routes from the sender-supplied HostRecord without verifying the ping's authenticity (CWE-345), so a forged ping can silently reroute autonomous-vehicle telemetry and control traffic. A vendor patch exists (PR #123 / commit 2f5224d) and the flaw was reported by VulnCheck, but no public exploit or active exploitation has been reported.

Out-of-bounds heap read in DASYLab's .DSB file parser allows memory disclosure and potentially code execution when a victim opens a specially crafted workspace file. All DASYLab versions prior to 2026.0.0 are affected. The CVSS 4.0 vector assigns high confidentiality, integrity, and availability impact to the vulnerable system, indicating the flaw may be exploitable beyond simple information leakage - a heap OOB read during string conversion can sometimes be chained into controlled code execution depending on heap layout. No public exploit or CISA KEV listing is identified at this time.

Out-of-bounds memory read in DASYLab's .DSB file parser (all versions before 2026.0.0) allows an attacker to read beyond the bounds of an allocated data structure, potentially disclosing adjacent memory contents and crashing the application. Exploitation requires convincing a user - typically an engineer, lab technician, or researcher - to open a specially crafted DASYLab worksheet file. No public exploit code or active exploitation has been identified at time of analysis; vendor measX has released version 2026.0.0 as the fix.

Out-of-bounds heap read during .DSB file parsing in DASYLab (all versions prior to 2026.0.0) allows an attacker who can persuade a user to open a specially crafted data file to potentially disclose heap memory contents and crash the application. The root cause is CWE-125: insufficient validation of user-supplied length or offset fields during file-handling routines, causing reads a few bytes past an allocated heap buffer boundary. No public exploit code or active exploitation has been identified at time of analysis; vendor-released patch is version 2026.0.0.

Unauthenticated information exposure in Apache Allura through version 1.20.0 allows remote attackers to retrieve non-public project content via the search functionality, bypassing access controls enforced elsewhere in the application. The search subsystem fails to apply visibility restrictions consistently, meaning private tickets, wiki pages, discussions, or code artifacts indexed by search may be readable without authentication. No public exploit code has been identified at time of analysis, and the EPSS score of 0.20% (10th percentile) reflects minimal observed exploitation interest despite the unauthenticated attack vector.

Unauthenticated exposure of OAuth client_secret credentials via the /oauth2/register endpoint affects Medplum versions 4.1.10 through 5.1.6, a healthcare developer platform widely used to build FHIR-compliant apps. When a caller submits a redirect_uri that matches a pre-configured entry in the server's defaultOAuthClients configuration block, the registration endpoint reflects the stored client_secret back in the response - effectively handing an attacker the credential needed to impersonate that OAuth client. The vendor patched this in version 5.1.7 (PR #8837); no public exploit has been identified at time of analysis.

Token forgery in python-jose through 3.5.0 lets an attacker who possesses the service's asymmetric public key mint valid HS256 (HMAC) JWTs, because the library's HMAC key initialization accepts DER-encoded public keys that lack PEM armor or SSH prefixes and never rejects them as asymmetric material. This is a classic JWT algorithm-confusion (RSA/EC-to-HMAC) attack that succeeds whenever the verifying application does not explicitly restrict the accepted algorithms, and it represents an incomplete fix for CVE-2024-33663. There is no public exploit identified at time of analysis, though the technique is well-documented and the same class of flaw has prior public tooling.

TLS certificate validation is disabled by default in the Federator component of Eclipse aeriOS, enabling man-in-the-middle interception of all outbound HTTPS connections made by the component. Any attacker who can position themselves on the network path between the Federator and external services - including OAuth authorization servers - can impersonate those services and harvest OAuth client credentials and bearer tokens. This affects only the current pre-release development codebase; a source-level fix is confirmed in commit 9c63b60, but no official tagged release exists. No public exploit or CISA KEV listing has been identified at time of analysis.

Improper TLS certificate validation in MISP's CurlClient class allows an adjacent-network attacker to perform man-in-the-middle attacks against outbound HTTPS connections made by the MISP instance, including threat intelligence synchronization. The PHP property `CurlClient::$verifyPeer` was left uninitialized, defaulting to null, which cURL interprets as falsy and therefore disables peer certificate verification. A successful attacker can observe authentication material and exchanged threat intelligence, and inject falsified data into MISP's responses from remote services. No public exploit has been identified; SSVC rates exploitation as none at time of analysis.

Cross-tenant vault key disclosure in Helicone's AI proxy and observability platform exposes plaintext upstream provider credentials - including OpenAI, Anthropic, and AWS Bedrock API keys - to any authenticated admin or owner of a separate tenant organization. The vulnerability exists in VaultManager.getDecryptedProviderKeyById(), invoked via GET /v1/vault/key/{providerKeyId}, which decrypts and returns vault secrets without verifying that the requesting user's organization matches the organization that owns the target key. A vendor patch commit is available; no active exploitation or public proof-of-concept has been identified at time of analysis.

TLS private key disclosure in SiYuan <=3.8.1 allows any authenticated user - or any network client on default no-auth-code deployments - to exfiltrate the server's TLS private key (conf/key.pem) and CA private key (conf/ca.key) via a simple POST to /api/file/getFile. The root cause is a dual flaw: the IsForbiddenAbsPath() blocklist covers only conf/conf.json by exact match while leaving sibling key material unprotected, and the getFile handler unconditionally skips the blocklist for RoleAdministrator - a role granted to every authenticated user in v3.8.1. On TLS-enabled, network-exposed instances this enables retrospective decryption of captured HTTPS sessions and forgery of certificates already trusted by clients that imported SiYuan's CA. No public exploit identified at time of analysis; vendor-confirmed fix is available in v3.8.2.

SiYuan's full-text search logging mechanism writes admin API tokens in plaintext to an accessible log file when requests exceed internal timing thresholds, enabling authenticated low-privilege users to escalate to permanent administrative access. All self-hosted SiYuan instances before v3.8.2 are affected; because API tokens are passed as query parameters and the log file is retrievable via the built-in getFile endpoint, the privilege escalation path requires no additional exploitation steps beyond triggering a slow search and reading the file. No public exploit code has been identified at time of analysis, and the vulnerability is not listed in the CISA KEV catalog. Vendor-released patch v3.8.2 resolves the issue.

Out-of-bounds memory read in the Linux kernel's DRM HDMI connector subsystem allows a local, low-privileged user to potentially leak up to 32 bytes of adjacent kernel heap memory or trigger a page fault leading to kernel crash. The flaw exists in a helper function that copies audio infoframe data using the destination buffer size (60 bytes) rather than the source structure size (28 bytes), reading 32 bytes beyond the source allocation. No public exploit code exists, and EPSS sits at the 10th percentile; however, the availability impact is real - the overread can cross a page boundary and cause an access violation.

Uninitialized heap memory disclosure in FreeRDP 3.0.0-3.30.0 exposes up to 576 bytes of stale server or proxy heap contents - potentially including cleartext credentials from prior RDP sessions - to any authenticated downstream RDP client. Three PDU-serialization functions in libfreerdp/core/info.c advance the stream write pointer with Stream_Seek over reserved padding bytes instead of zeroing them with Stream_Zero, leaving prior malloc heap contents in Save Session Info PDUs transmitted to clients. FreeRDP-based server deployments and freerdp-proxy instances are the affected roles; a vendor-confirmed fix ships in version 3.31.0, and no public exploit or active exploitation has been identified at time of analysis.

HTTPX2's SOCKS5 proxy implementation omits TLS negotiation for WebSocket Secure (wss://) origins, exposing the full WebSocket session - opening handshake, query parameters, Authorization headers, cookies, and message frames - in plaintext to anyone who can observe or control the proxy path. Affected are applications using Client.websocket() or AsyncClient.websocket() in HTTPX2 versions 2.6.0 through 2.9.1 when routing through a SOCKS5 proxy. An adjacent-network attacker can passively eavesdrop on or actively modify traffic, and can impersonate the WebSocket server due to the missing certificate verification step. No public exploit code has been identified at time of analysis and this CVE does not appear in the CISA KEV catalog.

Verification bypass in prevail eBPF verifier versions before 0.2.4 allows any caller with eBPF program submission access - including unprivileged users on BPF-enabled kernels - to load programs that incorrectly pass safety checks but destroy the upper 32 bits of kernel pointer registers at runtime via ALU32 ADD and SUB instructions. The runtime effect is either a program fault (availability impact) or exposure of the lower 32 bits of a kernel pointer as a leakable scalar, partially defeating kernel ASLR. A vendor patch is available at v0.2.4; no active exploitation or public exploit code has been identified at time of analysis.

Verifier bypass in PREVAIL (vbpf/prevail), the polynomial-runtime eBPF static verifier, before version 0.2.4 lets an attacker get an unsafe eBPF program accepted as safe. The abstract transformer treats stores through a context-typed (T_CTX) base register as a silent no-op, so a program can overwrite a context pointer field such as ctx->data, reload it typed as a packet pointer (T_PACKET), and dereference an attacker-controlled address while prevail still reports PASS, enabling out-of-bounds memory access. Fixed in 0.2.4; no public exploit identified at time of analysis beyond the maintainer's regression test cases, which concretely demonstrate the bypass shape.

Unauthenticated sensitive data exposure in dotstore's WooCommerce Product Attachment plugin (versions <= 2.3.3) allows remote, unauthenticated attackers to retrieve product attachment files that should be access-controlled. The vulnerability stems from insufficient authorization checks on attachment retrieval endpoints, classified under CWE-497 (Exposure of Sensitive System Information to an Unauthorized Control Sphere). No active exploitation has been confirmed by CISA KEV, and no public exploit code is identified at time of analysis, but the low attack complexity and zero-authentication requirement make opportunistic access trivial for any network-reachable attacker.

Cross-user session confusion in libcurl allows one user's HTTP request to be silently transmitted over another user's previously authenticated Negotiate (SPNEGO/Kerberos) connection. When an initial request to a host is made with empty credentials, libcurl incorrectly treats the resulting connection as eligible for reuse, so a subsequent request from a different principal inherits the earlier authenticated context. There is no public exploit identified at time of analysis; EPSS is low (0.26%, 17th percentile), and the flaw is only meaningful in multi-user or proxy-style deployments that share a single libcurl handle/connection pool.

Secure-flag stripping in curl 8.13.0 through 8.21.0 allows a server to cause curl to transmit session cookies over unencrypted HTTP by returning a Set-Cookie header that places a horizontal tab character (ASCII 9) instead of a space immediately before the Secure attribute. curl's cookie parser fails to recognise the attribute under this condition, stores the cookie without its Secure constraint, and subsequently includes it in plaintext HTTP requests to the same host. No public exploit code has been identified and EPSS probability is 0.22% (12th percentile), reflecting low observed exploitation activity despite the straightforward triggering mechanism.

Use-after-free in libcurl's HTTP/2 Server Push handling affects applications that enable server push while sharing connections across easy handles via the curl share interface. A malicious or compromised HTTP/2 server can trigger the freed-then-reused memory during connection cleanup, leading to memory corruption, crashes, and potential integrity impact against any of the vast range of software that embeds libcurl. EPSS is low (0.21%, 11th percentile) and there is a referenced HackerOne report (#3916059) but no public exploit code and no CISA KEV listing; no public exploit identified at time of analysis.

curl and libcurl through version 8.21.0 incorrectly handle domain-scoped cookies for Public Suffix List (PSL) domains, permitting a malicious server to set cookies scoped to a PSL entry rather than rejecting them as standards require. This flaw enables cross-subdomain cookie propagation under shared public suffix registries (e.g., github.io, co.uk), potentially exposing session tokens or enabling session fixation against other services the victim connects to under the same PSL domain. No public exploit code exists and CISA KEV listing is absent; the vulnerability was fixed in curl 8.22.0 released September 2, 2026.