Regular Expression Denial of Service in SAP S/4HANA (Private Cloud) allows unauthenticated remote attackers to exhaust server resources by supplying specially crafted input to a network-accessible endpoint that invokes a vulnerable third-party regex component (CWE-1333). Successful exploitation causes high availability impact with no confidentiality or integrity exposure, making this a targeted disruption risk against a critical ERP platform. No active exploitation has been confirmed - the vulnerability is not listed in CISA KEV and no public exploit code has been identified at time of analysis.
Remote code execution outside the browser sandbox is achievable in Google Chrome prior to 152.0.7977.65 via a race condition in ANGLE, the graphics API translation layer. Exploitation requires a victim to visit a crafted HTML page (UI:R), and the high attack complexity (AC:H) reflects the timing-sensitive, probabilistic nature of race condition exploitation. No public exploit code has been identified and CISA KEV does not list this CVE, though SSVC rates technical impact as total if exploitation succeeds.
Heap-based buffer overflow (CWE-122) in Blink, Chrome's rendering engine, grants native code execution within the renderer sandbox to attackers who have already achieved renderer process compromise, affecting all Chrome versions prior to 152.0.7977.65 across all platforms. The fix is confirmed in stable channel release 152.0.7977.65. No public exploit identified at time of analysis; EPSS at 0.28% (20th percentile) and Chromium's own Medium severity rating both reflect the high attack complexity imposed by the mandatory prior-renderer-compromise precondition.
Sandbox escape in Google Chrome's Media component on Windows allows a remote attacker who has already compromised the renderer process to execute arbitrary code outside Chrome's sandbox via a crafted HTML page. Affected versions are all Chrome releases on Windows prior to 152.0.7977.65. No public exploit code or active exploitation has been identified at time of analysis; SSVC assessment confirms exploitation status as none, and EPSS probability is low at 0.26% (17th percentile), though SSVC technical impact is rated total given full sandbox bypass capability.
Stored cross-site scripting in YOOtheme Pro Extension for Joomla versions 1.0.0 through 5.0.41 allows authenticated, privileged users to inject persistent malicious scripts via the location custom field, which lacks HTML output escaping before rendering user-supplied content in the browser. Any site visitor who subsequently loads a page rendering the tainted field will execute the attacker-controlled script in their browser context, enabling session hijacking, credential theft, or unauthorized actions on their behalf. No public exploit code or active exploitation has been identified at time of analysis, and the requirement for a high-privilege Joomla account substantially limits the threat surface.
Remote code execution within the Chrome sandbox is achievable via a race condition in the V8 JavaScript engine, affecting all Google Chrome versions prior to 152.0.7977.65. An unauthenticated remote attacker (per CVSS PR:N) can trigger the vulnerability by luring a user to a crafted HTML page, exploiting a timing window in V8's memory management to execute arbitrary code. No public exploit code or active exploitation has been identified at time of analysis; EPSS sits at 0.24th percentile and CISA SSVC records exploitation status as none, classifying this as a theoretical-but-patched risk for organizations that deploy promptly.
Uncontrolled resource consumption in Apache Tomcat's HTTP/2 implementation allows remote unauthenticated attackers to exhaust server memory by repeatedly opening and resetting HTTP/2 streams, exploiting an allocation leak in the backlog tracking mechanism that fails to release memory on stream reset. Affected branches span Tomcat 9.0.39-9.0.120, 10.1.0-M1-10.1.57, and 11.0.0-M1-11.0.24, with EOL branch 8.5.59-8.5.100 also confirmed vulnerable but unpatchable. No public exploit has been identified at time of analysis; EPSS stands at a low 0.22%, but the unauthenticated network vector and high availability impact make this operationally significant for any HTTP/2-enabled Tomcat deployment.
Cache poisoning via Host header injection in WP Fastest Cache before version 1.5.1 exposes every WordPress site visitor to unauthenticated JavaScript execution from attacker-controlled servers. The plugin incorporates the client-supplied HTTP Host header into asset file URLs baked into cached pages while omitting that header from the cache key, meaning a single forged request can persist malicious script references across all subsequent page serves. No public exploit code has been identified at time of analysis, but the technique - HTTP Host header injection leading to web cache poisoning - is well-documented and the attack carries a CVSS 3.1 score of 7.5.
Sandbox escape in Google Chrome's Media component (prior to 152.0.7977.65) allows a remote attacker who has already compromised the renderer process to execute arbitrary code outside the sandbox via a crafted HTML page. Google has released a patched stable channel build (152.0.7977.65), with no public exploit code or active exploitation confirmed at time of analysis. EPSS sits at 0.21% (12th percentile) and SSVC rates the exploitation status as none with automatable: no, indicating this is a low-immediacy but technically severe sandbox escape requiring a two-stage attack chain.
Rewrite valve processing in Apache Tomcat contains an off-by-one error (CWE-193) in the handling of the [N] (Next) flag, causing rule iteration to restart from the second rewrite rule instead of the first, which can produce unintended looping behavior and result in a denial-of-service condition. Affected supported versions span Tomcat 9.0.0.M1 through 9.0.120, 10.1.0-M1 through 10.1.57, and 11.0.0-M1 through 11.0.24; the EOL 8.5.x branch is also affected but will not receive a fix. No active exploitation is confirmed (not in CISA KEV), EPSS stands at 0.20% (10th percentile), and no public exploit code has been identified at time of analysis.
Local file disclosure in the Kaltura HTML5 player's mwEmbedLoader.php endpoint allows unauthenticated remote attackers to read arbitrary files from the server's filesystem. The root cause is that the ServiceUrl parameter - used as a base URL for backend requests - accepts non-HTTP URI schemes including file://, which the server then fetches; when an exception is raised, the raw deserialized response is reflected back to the client in the error message, leaking file contents. Affected releases span html5lib v2.45, v2.103, and earlier v2.x versions; no public exploit or KEV listing has been identified at time of analysis, though the attack path is straightforward for any attacker with access to the exposed endpoint.
Race condition in Google Chrome Enterprise on Windows prior to version 152.0.7977.65 enables an adjacent network attacker to potentially escape the browser sandbox and execute arbitrary code on the host system via crafted network traffic. No public exploit code has been identified and CISA has not listed this in the Known Exploited Vulnerabilities catalog. Despite a CVSS base score of 7.5, Google's Chromium team assigned an internal severity of Low, and SSVC analysis independently rates exploitation as none with non-automatable attack conditions - together suggesting constrained real-world impact relative to raw score.
Observable timing discrepancy in the Drupal Token Content Access module exposes token-protected content to brute-force enumeration via side-channel analysis. All module versions from 0.0.0 through 3.1.2 are affected; unauthenticated remote attackers (PR:N, AV:N per CVSS) can measure differential response times during token validation to reconstruct valid access tokens character-by-character, ultimately bypassing content access restrictions. No active exploitation has been confirmed (SSVC exploitation: none, EPSS 0.14%, no KEV listing), and a vendor patch is available per Drupal SA-CONTRIB-2026-090.
DPoP proof replay attack in Spring Security's DPoPProofJwtDecoderFactory allows unauthenticated network attackers to bypass token binding protections by evicting cached JWT ID entries through request flooding, then replaying captured valid DPoP proofs to impersonate legitimate users. Affected versions span Spring Security 6.5.0-6.5.11, 7.0.0-7.0.6, and 7.1.0, impacting any Java application using Spring Security's DPoP-based OAuth 2.0 validation. No public exploit code or CISA KEV listing has been identified at time of analysis, but the CVSS C:H/I:H ratings reflect full token-holder impersonation upon successful exploitation.
Authentication bypass in Apache HiveServer2 4.0.0-4.2.0 allows unauthenticated network attackers to impersonate any Hive user and obtain a fully authenticated HiveServer2 session when SAML mode is enabled over HTTP transport. The root cause is an inverted HMAC signature comparison in HiveSamlAuthTokenGenerator.java - the original code used !MessageDigest.isEqual(), causing all tokens to pass validation regardless of signature correctness, meaning a forged bearer token with any embedded username is unconditionally accepted. No CISA KEV listing and an EPSS of 0.14% (4th percentile) indicate no confirmed active exploitation at time of analysis, but the logic flaw is trivially exploitable once the deployment condition is met.
CustomTabs in Google Chrome for Android prior to version 152.0.7977.65 exposes a missing authorization flaw (CWE-862) that allows a local attacker, operating through a co-installed malicious application, to potentially escape the Chrome sandbox and execute arbitrary code. The Chromium security team rated this Medium severity internally, and the EPSS score of 0.13% at the 3rd percentile confirms no observed exploitation activity. A vendor patch has been released in Chrome 152.0.7977.65; no public exploit code or CISA KEV listing exists at time of analysis.
Command injection in NVIDIA UFM Enterprise's IBDiagnet API allows an authenticated administrator with adjacency-level network access to execute arbitrary OS commands on the host system, potentially achieving full compromise. All major release trains are affected - GA, LTS 2023, LTS 2024, and LTS 2025 - as confirmed by NVIDIA's own advisory and NVD CPE data. No public exploit code or CISA KEV listing has been identified at time of analysis, keeping real-world opportunistic risk constrained by the high privilege and network adjacency prerequisites.
Local privilege escalation to root is possible in Faktory background job server versions before 1.10.0 through an insecure temporary file flaw in its embedded Redis bootstrapper. The bootstrapper writes Redis startup configuration to the world-writable path /tmp/redis.conf and never validates or overwrites the file on subsequent boots, allowing a local unprivileged user to pre-plant a malicious config before Faktory starts. Because the official systemd unit runs both Faktory and its Redis child process as root, an attacker-supplied loadmodule directive in that config enables arbitrary native code execution as root. No public exploit has been identified at time of analysis, and the issue is fixed in version 1.10.0.
Authentication bypass in LACT (Linux AMD GPU Configuration Tool) through version 0.10.0 allows a local attacker with low privileges to circumvent Polkit-based access controls by exploiting PID reuse in the UnixProcessSubject / Peer PID authentication mechanism (CWE-290). By winning a race condition between Polkit's process identity check and authorization grant, an attacker can spoof their process as a legitimately authorized caller and gain unauthorized control over privileged GPU configuration operations. No public exploit code has been identified and no active exploitation has been confirmed at time of analysis.
Arbitrary file deletion in Adminer before 5.4.3 exposes any file writable by the PHP process to authenticated attackers operating in SQLite mode by abusing unvalidated relative path input in the db[] parameter during database drop operations. The root cause is a missing file extension check in the SQLite drop action (CWE-22 path traversal), allowing directory traversal sequences to escape the intended SQLite database directory. No public exploit code and no CISA KEV listing have been identified at time of analysis; however, the deterministic nature of path traversal exploitation makes the vulnerability straightforward to weaponize once authenticated access is obtained.
Privilege escalation in ShopEngine (all versions up to and including 4.9.4) allows Shop Manager-level authenticated users to achieve full WordPress site takeover by injecting arbitrary core options through a malicious WXR import file. The plugin's `rum_importer()` function is hooked to the WordPress core `import_start` action without any plugin-owned capability check or allowlist filtering, meaning attacker-controlled `<wp_option>` pairs from a crafted import file flow directly into `update_option()` - enabling manipulation of `users_can_register` and `default_role` to open administrator self-registration. No CISA KEV listing or EPSS score was present in the provided data; exploitation requires an existing Shop Manager account and WooCommerce's grant of the `import` capability to that role, but once those conditions are met the path to full site compromise is straightforward and documented.
Stored Cross-Site Scripting in the Forminator Forms WordPress plugin (all versions through 1.57.0.2) allows unauthenticated attackers to inject persistent JavaScript payloads that auto-execute in the WordPress admin panel when an administrator views the Submissions page. The attack vector is the Save-and-Continue draft AJAX endpoint, which is registered without privilege checks (nopriv), bypassing radio field option-membership validation and permitting arbitrary values to be stored. Payload execution is triggered automatically via the bundled Inputmask library's data-attribute callback binding, requiring no further attacker interaction after initial injection. No public exploit code or CISA KEV listing is identified at time of analysis, though the unauthenticated, zero-complexity attack path and high-value admin-panel target represent meaningful real-world risk.
DOM-based reflected XSS in the Forminator Forms WordPress plugin (all versions up to and including 1.57.0) allows unauthenticated network attackers to execute arbitrary JavaScript in victim browsers by embedding a payload in the 'error_description' URL parameter, which the plugin's front.stripe.js writes to the DOM without sanitization. The attack surface expanded significantly when Stripe Checkout Sessions became the plugin's default payment method in version 1.56.0, meaning any site that upgraded to 1.56.0 or 1.57.0 without changing payment settings is exposed on pages hosting a Forminator form. No public exploit has been identified at time of analysis, but the vulnerable code paths at four specific locations in front.stripe.js are publicly documented in Wordfence's disclosure, with the fix released in version 1.57.0.1.
Scope enforcement in Ech0 4.3.4 and earlier is opt-in at the route level, meaning privileged admin routes that omit the RequireScopes() middleware fall back to role-only checks and honor any valid admin token regardless of declared scope. An admin holding a deliberately narrowed API token - for example one scoped to echo:read - can reach endpoints such as /api/inbox, /api/panel/comments, and /api/backup/export, with the backup handler additionally stripping scope metadata entirely and returning a full database ZIP archive. No CISA KEV listing or confirmed public exploit code has been identified; the vulnerability is fixed in an upstream release (version discrepancy noted below).
Stored Cross-Site Scripting in NotificationX Pro for WordPress (all versions through 3.1.4) allows unauthenticated remote attackers to persistently inject arbitrary JavaScript into site pages, executing in any visitor's browser context. The CVSS vector (AV:N/AC:L/PR:N/UI:N/S:C) reflects the combination of zero-privilege access and a scope change impact reaching beyond the WordPress installation itself into the browser sessions of site visitors. No public exploit code or CISA KEV listing has been identified at time of analysis, but the unauthenticated injection surface and persistent payload storage make this a credible risk for any site running the affected plugin.
Arbitrary file deletion in Grav CMS before 2.0.16 allows authenticated users holding media management permissions to delete files anywhere on the server filesystem via path traversal sequences in filenames passed to MediaUploadTrait::deleteFile(). The root cause is that deleteFile() validates only the basename of a supplied filename while passing the full, unvalidated path - including ../ sequences - directly to PHP's unlink(). No public exploit or CISA KEV listing exists at time of analysis, but the low attack complexity makes this straightforward to weaponize for any user who holds media upload/delete rights.
Uncontrolled ZIP extraction in LeafWiki 0.10.0 through 0.12.0 allows any authenticated Editor or Admin to exhaust server disk or memory by uploading a crafted ZIP bomb via the import route. The vulnerable function ZipExtractor.ExtractToDir uses Go's io.Copy without a decompressed-size limit, so a maximally compressed archive well inside the 500 MiB upload cap can expand to hundreds of gigabytes on disk - or into RAM when the OS temporary directory is tmpfs-backed. No public exploit has been identified at time of analysis, though the attack is mechanically simple for any credentialed Editor.
Unrestricted PHP file upload in Adminer's AdminerFileUpload plugin (versions before 5.4.3) allows any authenticated low-privilege user to deploy a webshell and execute arbitrary code as the web-server user. The root cause is a permissive default extension allowlist that fails to block .php uploads to columns with _path suffixes when the uploadPath directory is web-served. No public exploit has been identified at time of analysis, but the attack chain is technically trivial for anyone with Adminer credentials and requires no user interaction.
Out-of-bounds pointer arithmetic in OpenEXR's TypedDeepImageChannel::row() - and parallel methods in FlatImageChannel and SampleCountChannel - affects all releases before 3.2.11, 3.3.13, and 3.4.14, exposing applications that process deep EXR files with nonzero dataWindow origins to deterministic crash or limited heap memory disclosure. The CVSS 7.1 score (AV:N/AC:L/PR:N/UI:R) reflects unauthenticated, low-complexity delivery of a crafted image file requiring only that a user or pipeline opens it, with high availability impact and limited confidentiality exposure. No CISA KEV listing and no public exploit code have been identified at time of analysis; patched releases are available from the Academy Software Foundation.
Denial-of-service vulnerabilities in rclone's archive backend SquashFS parser allow an attacker with write access to rclone-exposed storage to crash the rclone process or exhaust CPU. Affected versions v1.72.0 through v1.74.4 fail to validate attacker-controlled superblock and inode metadata values sourced from the go-diskfs dependency before use, enabling three distinct panic/loop conditions. When the :archive: remote is exposed via 'rclone serve sftp', exploitation crashes the entire SFTP server. No public exploit has been identified at time of analysis, but the attack surface is concrete and parser triggering is straightforward once a malicious image is placed.
Path traversal in Grav CMS before 2.0.16 exposes arbitrary server-side files to authenticated page authors through the `media_directory()` Twig function, which fails to validate attacker-supplied filesystem paths. Users holding page authoring privileges can supply arbitrary paths to `media_directory()` and then invoke the allow-listed filepath accessor on the resulting Medium object to read the contents of any file accessible to the web server process whose extension matches the CMS media allow-list. No public exploit has been identified at time of analysis and the vulnerability is not listed in CISA KEV, but the low privilege requirement and high confidentiality impact make this a meaningful risk in any multi-author or externally-accessible Grav deployment.
Unauthorized organizer attribution in the TYPO3 Events 2 frontend management extension allows any authenticated organizer-role user to create events falsely attributed to another organizer's record by supplying a foreign organizer identifier in the event creation request. The flaw stems from a permission check that validates only the presence of any organizer role rather than verifying the submitting user's ownership of the specific organizer referenced in the request. No public exploit has been identified and this vulnerability is absent from CISA KEV; however, the CVSS 4.0 score of 7.1 reflects a high integrity impact on the vulnerable system driven by straightforward, low-privilege network exploitation with no additional attack requirements.
Server-side request forgery in Dradis Community Edition 5.1.0-5.2.0 allows any authenticated non-admin user to hijack the AI provider configuration and redirect server-initiated HTTP requests to arbitrary URLs, including internal cloud metadata endpoints such as 169.254.169.254. A broken authorization gate in ProvidersController and AgentsController is conditioned on the Ruby constant Dradis::Pro, which is never defined in CE builds, so the admin_required filter is silently skipped for all CE installations. Non-2xx responses from attacker-supplied URLs are reflected verbatim to the attacker's browser via ActionCable/Turbo Stream, enabling exfiltration of internal service content. No public exploit code or CISA KEV listing has been identified at time of analysis.
Missing administrator authorization on dashboard log endpoints in Ech0 before 4.4.3 allows any authenticated low-privilege user to read historical system logs and subscribe to live log streams via GET /api/system/logs, SSE, and WebSocket endpoints. Affected deployments expose operational data including internal file paths, stack traces, admin activity records, background job output, and internal URLs to any valid session holder. No public exploit has been identified at time of analysis, and this CVE is not listed in the CISA KEV catalog, but the low exploitation complexity and high confidentiality impact make this a meaningful post-authentication information disclosure risk for multi-user deployments.
Out-of-bounds read in OpenEXRUtil's SampleCountChannel::row() API allows an attacker to crash applications or leak adjacent heap memory by supplying a crafted deep EXR file with a non-zero dataWindow origin. Affected are OpenEXR releases through 3.2.10, 3.3.0-3.3.12, and 3.4.0-3.4.13 - the complete release history prior to the coordinated January 2026 patch. The CVSS vector (AV:N/AC:L/PR:N/UI:R) reflects that exploitation requires no authentication but does require a user or automated pipeline to open an attacker-controlled file; no public exploit code and no CISA KEV listing exist at time of analysis.
Unrestricted file upload in ContiNew Admin through v4.1.0 allows authenticated users to bypass server-side validation entirely on multipart upload endpoints, storing files with arbitrary extensions in the backend storage. The flaw stems from missing permission checks and the absence of any file-type allowlist in MultipartUploadController.java, meaning a low-privileged account holder can upload content such as server-side scripts. Uploaded files are then accessible via web server URLs, creating a direct path to webshell deployment or persistent malicious content delivery. No public exploit or CISA KEV listing is confirmed at time of analysis.
Heap out-of-bounds read in OpenEXRUtil's deep and flat image channel row() API allows an attacker supplying a crafted EXR file with a non-zero dataWindow origin to crash any application that processes the file, with conditional heap memory disclosure dependent on attacker-controlled heap layout. Affected are OpenEXRUtil versions 3.3.0-3.3.12 and 3.4.0-3.4.12; the flaw exists across three channel header implementations (ImfDeepImageChannel.h, ImfFlatImageChannel.h, ImfSampleCountChannel.h). No public exploit identified at time of analysis, and patched releases 3.3.13 and 3.4.13 are available.
Arbitrary file read in GitPython before 3.1.59 allows low-privileged remote attackers to exfiltrate sensitive files by supplying a crafted reference argument to TagReference.create(). The vulnerability arises because a positional reference parameter (e.g., '--file=/etc/passwd') bypasses the library's built-in unsafe option guard, causing git to interpret the value as a flag and return the targeted file's contents embedded in the annotated tag message. No public exploit code or CISA KEV listing has been identified at time of analysis, but the CVSS 4.0 score of 7.1 with high confidentiality impact reflects genuine disclosure risk in any application that exposes this code path to user-controlled input.
Configuration secret disclosure in Grav CMS before 2.0.16 allows authenticated content editors to extract sensitive values - including cache credentials - from system, site, and theme configuration arrays via the Twig templating engine. The sandboxed Twig render context fails to filter these configuration objects, permitting attackers to bypass the intended config_denied_paths restrictions using standard dot-notation syntax. No public exploit code or CISA KEV listing has been identified at time of analysis, though the CVSS 4.0 score of 7.1 reflects a high confidentiality impact for installations where editor access is not fully trusted.
Arbitrary file read in GitPython versions before 3.1.59 exposes sensitive filesystem content to low-privileged network attackers through an incomplete argument-injection guard. The unsafe_git_revision_options denylist in Repo.blame() omits the --contents and -S git options, allowing attackers to supply revision values such as --contents=/etc/passwd and receive the target file's contents embedded in the blame output. No public exploit has been identified at time of analysis, but the bypass is conceptually trivial - a crafted string is all that is required once access to the blame API is established.
Heap out-of-bounds write in OpenEXR's TiledRgbaInputFile RGBA API affects all three active release lines on 32-bit/ILP32 builds, allowing a remote attacker to deliver a crafted EXR file that crashes or corrupts the heap of any application that opens it. The root cause is a 32-bit integer overflow in the Array2D<Rgba> buffer size calculation when tile dimensions (65537×65537) are encoded in the file header but the actual dataWindow is tiny (40×40), causing the allocator to produce an undersized buffer that tile decode then overflows. No public exploit code has been identified and the vulnerability is not listed in CISA KEV; vendor-released patches are available across all three affected release branches.
Out-of-bounds and use-after-free memory writes in OpenEXR's FlatImageChannel row() API expose rendering pipelines and image-processing services to denial-of-service and potential memory corruption when handling untrusted EXR files. Affected versions include all releases before 3.2.11, 3.3.0 through 3.3.12, and 3.4.0 through 3.4.13; the flaw is triggered specifically when an application calls FlatHalfChannel::row() on an EXR image with a nonzero dataWindow.min value. No active exploitation has been confirmed (not listed in CISA KEV), and no public exploit code has been identified at the time of analysis.
Heap out-of-bounds write in OpenEXR 3.3.0-3.3.12 and 3.4.0-3.4.13 allows an attacker who can supply a crafted deep scanline EXR file to crash the exrmetrics utility or potentially execute arbitrary code when pixel conversion options are active. The flaw arises from a HALF/FLOAT element-size mismatch: sample buffers are allocated for 2-byte HALF input while DeepSlice writes 4-byte FLOAT output, overrunning the heap. No public exploit or active exploitation has been identified at time of analysis; fixed releases 3.3.13 and 3.4.14 are available.
Cross-organizer event modification in the TYPO3 Events 2 extension allows any authenticated frontend event manager to tamper with events owned by other organizers. The frontend management update flow performs its authorization check against a different event identifier than the one the request ultimately modifies - a textbook CWE-639 IDOR mismatch - granting write access beyond the user's legitimate scope. No public exploit has been identified at time of analysis, but with a CVSS 4.0 score of 7.1 (VI:H) and only low-privilege authentication required (PR:L), the integrity risk is meaningful in any multi-organizer TYPO3 deployment.
Heap buffer overflow in PyOpenEXR versions 3.3.0-3.3.12 and 3.4.0-3.4.13 allows an attacker to corrupt process memory by supplying a crafted EXR image file containing conflicting literal and prefixed RGB channel names. When the Python wrapper decodes such a file under the default separate_channels=False mode, a channel-name key collision causes the coalescing logic to reuse an undersized NumPy array and write pixel data out of bounds on the heap, producing high integrity and availability impact with no confidentiality exposure. No public exploit code has been identified and no active exploitation is confirmed; patched releases 3.3.13 and 3.4.14 are available from the Academy Software Foundation.
Heap out-of-bounds write in OpenEXR's exrmultiview utility allows an attacker who can supply crafted EXR files to corrupt heap memory when the tool combines two individually valid scanline EXR files whose union dataWindow is misaligned to one view's channel subsampling. Affected are OpenEXR versions across three maintained branches (pre-3.2.11, 3.3.0-3.3.12, and 3.4.0-3.4.13), all fixed in their respective patch releases. No public exploit or CISA KEV listing has been identified at time of analysis, though the attack is achievable via normal command-line invocation with attacker-supplied files in VFX pipeline contexts.
Arbitrary file write in sublinear-time-solver's MCP export_state and import_state tools allows any caller able to invoke the MCP interface to overwrite files accessible to the server process - including sensitive targets such as ~/.ssh/authorized_keys or application binaries - by supplying a path-traversal filepath argument. Affected packages are consciousness-explorer (npm) < 1.1.2, sublinear-time-solver (npm) < 1.6.0, and the Rust crate sublinear < 0.2.0. No public exploit or CISA KEV listing has been identified at time of analysis, but the write-anywhere primitive carries inherent high-impact potential against deployments that expose the MCP server to untrusted callers.
Incorrect authorization in the Chromoting component of Google Chrome on Windows prior to version 152.0.7977.65 enables a local attacker with standard user privileges to bypass system access restrictions by executing a crafted local program. The flaw (CWE-863) undermines Chrome Remote Desktop's authorization enforcement on Windows, allowing unauthorized access to protected system resources with both high confidentiality and integrity impact. No public exploit has been identified at time of analysis, and the EPSS score of 0.11% places this in the 1st percentile, indicating very low current exploitation probability despite the 7.1 CVSS base score.
OS command injection in zackees/transcribe-anything (versions up to 4.1.0) allows remote attackers to execute arbitrary system commands by supplying a crafted URL argument to the `ytdlp_download` function in `src/transcribe_anything/ytldp_download.py`. The CVSS 4.0 score is 2.9 - low - primarily because attack complexity is rated High, meaning exploitation requires specific conditions to align rather than trivial one-shot delivery. A publicly available proof-of-concept exploit exists via GitHub issue #137, and the vendor has not responded to the coordinated disclosure, leaving no patch available at time of analysis.
Unauthenticated OAuth redirect URI hijacking in Coroot versions 1.20.2 through 1.24.5 allows attackers to register malicious MCP OAuth clients without any credentials, then capture authorization codes from signed-in users who follow crafted authorization URLs and approve a consent dialog. The captured codes can be exchanged for valid access tokens, enabling full MCP session hijacking. No public exploit has been identified at time of analysis and no CISA KEV listing is present, but the registration step requires no authentication, materially lowering attacker prerequisites.