The internet is broken right now.
47 critical CVEs without a fix in the last 48 hours. Somewhere a staging server is public. It always is. 72 more already have a fix that someone still has to apply.
-
CVE-2026-119565
Plausible Analytics through 3.2.1 contains a mass assignment vulnerability in update_traffic_change_notification that allows authenticated site editors to reassign notifications by supplying a site_id parameter. Attackers can submit traffic_change_notification[site_id] set to another site's primary key via PUT /sites/:domain/traffic-change-notification/:type to receive that site's traffic alerts and visitor figures.
-
CVE-2026-130733
Grav from 2.0.0 before 2.2.18 contains an authorization bypass vulnerability that allows publishers to overwrite pages their per-page permissions deny by abusing FolderStorage key delimiter collisions and PageStorage folder reconstruction. Attackers can rename a folder like locked*** or save from a page with an unusual prefix width through a front-end Flex pages form, replacing protected content and permissions to access the protected subtree.
-
CVE-2026-111816
Apache Traffic Server forwarded a malformed 302 response that should have been rejected as a parse error, because the Transfer-Encoding: chunked carve-out bypassed the rejection path. A response that two peers can read differently is then stored and served, allowing cache poisoning and response-framing desync. This issue affects Apache Traffic Server: from 8.0.0 through 8.1.11, from 9.0.0 through 9.2.15, from 10.0.0 through 10.1.4, 10.2.0. Users are recommended to upgrade to version 9.2.16, 10.1.5, or 10.2.1, which fix the issue.
-
CVE-2026-115538
Moby's overlay network driver does not restore the iptables rules for encrypted overlay networks after firewalld deletes them. Unlike other Docker Engine networking components, the overlay driver does not handle firewalld reload or start signals over D-Bus. The missing rules cause outgoing VXLAN datagrams to bypass IPsec encryption and remove the protection against incoming cleartext VXLAN traffic. The condition is not logged by dockerd and persists until the encrypted network's last local endpoint is removed and another is added, or dockerd is restarted. Each further firewalld event repeats the condition.
-
CVE-2026-119833
Untrusted pointer dereference in Windows Message Queuing allows an authorized attacker to elevate privileges locally.
-
CVE-2026-85726
Chartbrew is an open-source web application that can connect directly to databases and APIs and use the data to create charts. Prior to 5.2.3, a date-typed MongoDB variable placed inside a quoted query placeholder was inserted as an unescaped string, unlike the safely escaped or coerced string, number, and Boolean branches. The vulnerable surface is mongoDB-backed public, shareable, or embedded charts processed by date-variable substitution. An unauthenticated user supplies a crafted URL filter to a public, shareable, or embedded MongoDB chart whose query places a date-typed variable in a quoted literal. The crafted date value breaks out of the quoted literal and changes query logic; the chart's returned row count becomes a Boolean oracle for character-by-character field extraction. The chart must be externally reachable and use the vulnerable date-in-quoted-placeholder branch. Extraction is read-only and limited to one queried collection per request, and it requires neither an account nor a known field value. The blind oracle can disclose any field in the queried collection, including secrets the chart does not display. This issue is fixed in version 5.2.3.
-
CVE-2026-57210
Heimdall is a cloud native Identity Aware Proxy and Access Control Decision service. Prior to version 0.17.17, when the trusted_proxies option is configured, heimdall extracts client IP addresses from the Forwarded (for= parameter) and X-Forwarded-For headers and exposes them as Request.ClientIPAddresses to the rule pipeline. However, extracted values are not validated to be syntactically valid IP addresses. Arbitrary strings, malformed IP literals, and RFC 7239 unknown values and obfuscated identifiers are accepted without further checks. In addition, the Forwarded header parser splits on , and ; without accounting for RFC 7239 quoted strings, which can cause a single quoted value to be parsed as multiple entries, with fragments — including trailing quote characters — treated as independent addresses. This issue has been patched in version 0.17.17.
-
CVE-2026-58705
In handleEvent of gscan.cpp, there is a possible way to trigger an out of bounds write on the heap due to a missing bounds check. This could lead to remote escalation of privilege with no additional execution privileges needed. User interaction is not needed for exploitation.
-
CVE-2026-86532
The bundled JAAS login modules leak whether a username exists and compare credentials in a non-constant-time manner. A remote unauthenticated attacker can enumerate valid accounts and may infer credential material through timing analysis. Users are recommended to upgrade to version 10.3.0 or 11.0.0-M2, which fix the issue.
-
CVE-2026-125106
Wagtail is an open source content management system built on Django. Prior to 7.0.9, 7.3.4, 7.4.3, and 8.0rc2, a Wagtail CMS user with the submit translations permission can call the Admin API copy for translation action for a source page that the user cannot edit. The action copies the inaccessible page into a translation, allowing the user to view its contents despite the missing source-page edit permission. This issue is fixed in versions 7.0.9, 7.3.4, 7.4.3, and 8.0rc2.
-
CVE-2026-123819
Wagtail is an open source content management system built on Django. Prior to 7.0.9, 7.3.4, 7.4.3, and 8.0rc2, a CMS user who has add permission for a snippet model but lacks change and view permission for an existing snippet can access the snippet copy view and copy that source object. The copy operation reveals the contents of a snippet that the user is not authorized to view. This issue is fixed in versions 7.0.9, 7.3.4, 7.4.3, and 8.0rc2.
-
CVE-2026-114647
The MongoDB C# Driver does not reject replacement documents whose top-level field names begin with a dollar sign in FindOneAndReplace, unlike other replace operations. An application that builds such a document from user-supplied field names can have the server apply update operators instead of a full replacement. The effect is limited to the document matched by the application's own filter, whose contents can be modified in ways the application did not intend.
-
CVE-2026-120991
When a container blob upload was rejected, for example because it exceeded a package size quota, the cleanup removed the blob from storage even if the same content already existed and was used by other packages. On an instance with a package quota configured, a user who can upload container images and has the exact content of another owner's package file or image layer could make it undownloadable. The cleanup now only removes blobs that the failed upload created.
-
CVE-2026-35102
HCL AION is affected by an open redirect vulnerability in the navigation bar of the GenAI web interface (Navbar component). The switch-version action redirected the browser to a dynamically constructed URL (window origin + '/workspace') without validating the redirect destination. This may be used in phishing attacks, potentially resulting in unintended behavior or security impact under certain conditions.
-
CVE-2026-120329
Saving the configuration of the mail backend module writes submitted settings to the page TSconfig without restricting them to the configuration options of the module. Backend users with access to the mail module can write arbitrary page TSconfig.
-
CVE-2026-90861
Progress Software WS_FTP Server versions prior to 2024.0.7 and 2025.0.4 on Windows may expose a user's password in clear text to another administrator, who could use it to gain broader administrative access to the server.
-
CVE-2026-106468
The Code Snippets WordPress plugin before 4.1.0 does not verify network-level capabilities before honouring a network-scope parameter on its snippet import routes on multisite, allowing an administrator of a single subsite without network-management rights to create, overwrite and deactivate network-wide snippets belonging to the whole network.
-
CVE-2026-125124
compose-lint is a Docker Compose security linter that identifies dangerous misconfigurations before deployment. In version 0.31.0, compose-lint does not neutralize the Azure Pipelines ##vso[ logging-command opener on any output path. A pull request can place that opener in a Compose value or file name, allowing output from the package, pre-commit hook, or Docker image to issue logging commands when printed to an Azure Pipelines job log. The maintainer did not reproduce the behavior on an Azure Pipelines agent. The GitHub Action is not affected because it runs only on GitHub Actions, and releases before 0.31.0 are outside this advisory because they did not claim Azure Pipelines coverage. This issue is fixed in version 0.32.0.
-
CVE-2026-112136
A vulnerability has been identified in Mendix Studio Pro 11 (All versions >= V11.15.0 < V11.16.0), Mendix Studio Pro 11.12 (All versions >= V11.12.6 < V11.12.7). The affected products MCP Server does not properly restrict its HTTP listener to the loopback interface, leaving it potentially exposed to remote systems. This could allow an attacker with network access to interact with the MCP Server remotely.
-
CVE-2026-123556
ZcashFoundation Zebra from 4.5.0 before 7.0.0 contains a protection mechanism failure that bans honest peers by scoring mempool lock-time rejections at the ban threshold. Attackers can broadcast transactions locked until the current height right after each block, causing lagging nodes to ban honest relaying peers for 24 hours or until restart.
-
CVE-2026-123791
A vulnerability was detected in OpenBMB Internet of Agents up to 9d78e88e47160eb42ee1ee8538c05e1c2c47877a. This affects the function _primary_field_query of the file common/utils/milvus_utils.py of the component Agent Registry. The manipulation of the argument key results in improper neutralization of special elements in data query logic. The attack may be launched remotely. The exploit is now public and may be used. This product operates on a rolling release basis, ensuring continuous delivery. Consequently, there are no version details for either affected or updated releases. The project was informed of the problem early through an issue report but has not responded yet.
-
CVE-2026-119179
A flaw in the connection management layer of the MongoDB Node.js driver can leave a pooled network connection in the pool for reuse while data from an earlier operation is still outstanding on it. In applications that enable an optional client-side operation timeout, an unauthenticated party that can send ordinary requests to the application may cause database replies to be paired with the wrong requests. This may result in unintended disclosure of other users' records, incorrect operation results being returned, and failures of legitimate requests.
-
CVE-2026-118996
CallBackery versions before 0.62.0 for Perl send users their own password hash and the admin-only user note via getUserConfig. getUserConfig() returns the caller's whole cbuser row, including the stored password hash and the note field that the user interface shows only to administrators. getTableData() and getAllFieldValues() in the user management plugins also return the note to users without the admin right. A non-administrator can read the notes administrators keep about them, and an attacker who obtains a session token can get that user's password hash and crack it offline.
-
CVE-2026-111515
Dell Alienware Command Center (AWCC), versions prior to 6.14.105.0, contain an Improper Authorization in Handler for Custom URL Scheme vulnerability. An unauthenticated attacker with remote access could potentially exploit this vulnerability, leading to Command Execution and Launch of Phishing Attacks.
-
CVE-2026-127351
Vulnerability in the Oracle Hyperion Financial Close Management product of Oracle Hyperion (component: Reconciliation Manage). The supported version that is affected is 11.2.26.0.000. Easily exploitable vulnerability allows unauthenticated attacker with network access via HTTP to compromise Oracle Hyperion Financial Close Management. Successful attacks of this vulnerability can result in unauthorized access to critical data or complete access to all Oracle Hyperion Financial Close Management accessible data. CVSS 3.1 Base Score 7.5 (Confidentiality impacts). CVSS Vector: (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N).
-
CVE-2026-126316
Frappe HRMS from 16.0.0 before 16.20.1 contains an information exposure vulnerability that places the subscription key in the query string of outbound subscription status requests. Attackers with access to proxy, server, or monitoring logs along the request path can recover the site's subscription key from recorded URLs.
-
CVE-2026-93335
Vault Community Edition and Vault Enterprise did not consistently apply explicit ACL policy deny rules to request paths that differ only by a trailing slash. This may allow an authenticated user who holds a broader delete permission on system configuration paths to disable a secrets engine, auth method, or audit device that a more specific deny rule was intended to protect. This vulnerability (CVE-2026-93335) is fixed in Vault Community Edition 2.2.2 and Vault Enterprise 2.2.2, 1.21.14, and 1.19.25.
-
CVE-2026-96122
Untrusted pointer dereference in Windows Graphics Kernel allows an authorized attacker to elevate privileges locally.
-
CVE-2026-96129
Inclusion of functionality from untrusted control sphere in Windows Kernel allows an authorized attacker to elevate privileges locally.
-
CVE-2026-60044
The backups created by the application use a weak static password allowing an unauthenticated remote attacker to access their content. This can result in a total loss of confidentiality.
-
CVE-2026-119344
Cocos AI is a confidential computing system for AI. In version 0.8.2 and earlier, Cocos AI accepts trusted execution environment attestation Evidence when firmware patch levels and software measurements match, without binding the Evidence to an allowed platform owner or unique machine identity and without enforcing revocation. A remote attacker who controls or compromises another platform trusted by the same verifier can substitute valid Evidence from that platform and cause authorization of the wrong platform. This vulnerability is addressed in 0.9.0. However, this vulnerability can't be patched in source code alone, since the vulnerability arises from the confidential computing threat model itself.
-
CVE-2026-119210
MariaDB Connector/J is used to connect applications developed in Java to MariaDB and MySQL databases. From 3.0.0 until 3.3.7, 3.4.5, and 3.5.11, the ClientMessage.validateLocalFileName() check used a regular expression that could match LOAD DATA LOCAL INFILE text inside quoted identifiers, string literals, or comments instead of requiring a real statement. A malicious or compromised server, or a machine-in-the-middle attacker, could return crafted database or table identifiers that the connector reused in metadata or updatable-result queries, then answer with a LOCAL INFILE request for a client file. With allowLocalInfile enabled, this caused the connector to stream any readable client-host file named by the crafted text, including through DatabaseMetaData.getImportedKeys(), DatabaseMetaData.getExportedKeys(), or ResultSet.updateRow() paths. This issue is fixed in versions 3.3.7, 3.4.5, and 3.5.11.
-
CVE-2026-106648
The Royal Elementor Addons WordPress plugin before 1.7.1070 does not validate submitted form field values before substituting them into the recipient addresses of its form notification emails, allowing unauthenticated attackers to replace the recipient list and relay attacker-composed email from the site's own domain when a recipient field is configured to echo a submitted value.
-
CVE-2026-100164
Observable response discrepancy vulnerability in Rolantis Information Technologies Tourism Industry and Trade Co. Ltd. Agentis allows Account Footprinting. This issue affects Agentis: before 4.8.
-
CVE-2026-116201
Snipe-IT before 8.5.0 contains a SAML assertion replay vulnerability that allows attackers to reuse captured assertions by exploiting a non-atomic nonce check in the SAML acs handler. Attackers can send concurrent POST requests to /saml/acs with one captured assertion to obtain multiple authenticated victim sessions that evade IdP-driven revocation.
-
CVE-2026-127377
Vulnerability in the Oracle Hyperion Financial Close Management product of Oracle Hyperion (component: Reconciliation Manage). The supported version that is affected is 11.2.26.0.000. Easily exploitable vulnerability allows high privileged attacker with network access via HTTP to compromise Oracle Hyperion Financial Close Management. Successful attacks of this vulnerability can result in unauthorized creation, deletion or modification access to critical data or all Oracle Hyperion Financial Close Management accessible data as well as unauthorized read access to a subset of Oracle Hyperion Financial Close Management accessible data. CVSS 3.1 Base Score 5.5 (Confidentiality and Integrity impacts). CVSS Vector: (CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:U/C:L/I:H/A:N).
-
CVE-2026-115056
Vulnerability in the Oracle Hyperion Infrastructure Technology product of Oracle Hyperion (component: Installation and Configuration). The supported version that is affected is 11.2.26.0.000. Easily exploitable vulnerability allows high privileged attacker with network access via HTTP to compromise Oracle Hyperion Infrastructure Technology. While the vulnerability is in Oracle Hyperion Infrastructure Technology, attacks may significantly impact additional products (scope change). Successful attacks of this vulnerability can result in unauthorized access to critical data or complete access to all Oracle Hyperion Infrastructure Technology accessible data. CVSS 3.1 Base Score 6.8 (Confidentiality impacts). CVSS Vector: (CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:C/C:H/I:N/A:N).
-
CVE-2026-130921
Nonce reuse in WebRTC in Google Chrome prior to 158.0.8127.36 allowed a remote attacker to potentially obtain sensitive information via a crafted HTML page. (Chromium security severity: Low)
-
CVE-2026-77433
Kiwi TCMS is an open source test management system. From 16.1 until 16.2, the account confirmation page’s incomplete validation of the next parameter accepts protocol-relative URLs because the version 16.1 fix for CVE-2026-54724 only requires the value to start with a slash. An unauthenticated attacker can craft a confirmation URL on a trusted Kiwi TCMS hostname that redirects a user to an external site, enabling phishing and related downstream confidentiality or integrity impacts when the user follows the link. This issue is fixed in version 16.2.
-
CVE-2026-82840
Session fixation in Apache Wicket. org.apache.wicket.authroles.authentication.AuthenticatedWebSession is the authentication base class of Wicket's role-based security module, and its signIn(String, String) method is the entry point through which an application authenticates a session. Renewing the container session id at that point is the standard defence against an attacker who has fixed a session id in a victim's browser, and Wicket provides that renewal as org.apache.wicket.Session#changeSessionId(), whose documentation directs callers to invoke it on login for exactly that reason. AuthenticatedWebSession#signIn(String, String) calls it inside a guard that succeeds only on the first transition of the session's signed-in flag. The session id is therefore renewed when a session first signs in, and is not renewed when a session that is already signed in authenticates again. signIn(String, String) returns true either way, so neither the caller nor the application can tell that the renewal did not happen. An application is affected where it authenticates through AuthenticatedWebSession, which AuthenticatedWebApplication requires and which the shipped UsernamePasswordPanel uses, and where a session that is already signed in can authenticate again: a step-up prompt before a sensitive action, an account switch, or simply a sign-in form that stays reachable once the user is signed in. Nothing in the module hides that form from a signed-in session. Reaching the defect also requires an attacker to place a session id of their choosing in the victim's browser. JSESSIONID is set by the servlet container rather than by Wicket, so that needs a position Wicket does not grant, such as control of a host under the same registrable domain or a container configured to carry the session id in the URL, and the planted session has to be one the attacker has already signed in on, because a session that is not yet signed in does have its id renewed by the victim's authentication. An authentication attempt that fails signs the session out and closes the window. Applications that renew the session id in their own code after a successful authentication are not affected. As a workaround, renew the session id from the application's own code on every successful authentication, by calling Session#changeSessionId() at the end of the authenticate(String, String) implementation. This issue affects Apache Wicket: from 8.18.0 through 8.19.0, from 9.23.0 through 9.24.0, from 10.9.0 through 10.11.0. Users are recommended to upgrade to version 8.20.0, 9.25.0 or 10.12.0, which fix the issue.
-
CVE-2026-119298
croc is a tool that allows any two computers to simply and securely transfer files and folders. Prior to 11.5.2, the receive-side file-readiness loop in src/croc/croc.go treats any sender-controlled basename beginning with croc-stdin- as a trusted text-transfer artifact even when SendingText is false. This skips askReceiveOverwrite and rename handling for a predictable exact-name collision in the receiver's working directory, allowing a remote sender whose transfer the user accepts to overwrite or truncate a matching regular file through recipientInitializeFile, or to remove and replace a matching destination with an empty file or an in-directory symlink through createEmptyFileAndFinish, without confirmation. The impact is limited to pre-existing names in the croc-stdin- namespace that the sender can predict; other filenames retain the prompt, dotfiles outside that namespace are not reachable, and symlink targets remain confined to the receive directory. This issue is fixed in version 11.5.2.
-
CVE-2026-93006
A flaw was found in dnsmasq. The TFTP server can serve the wrong file contents because a recently used transfer cache is not invalidated when a transfer finishes. An attacker on the local network who can request a TFTP file may cause a later client to receive previously cached data instead of the requested file. In PXE network-boot environments this can substitute attacker-chosen boot content for a victim machine.
-
CVE-2026-117999
Race condition in the IPC component. This vulnerability was fixed in Firefox 159, Firefox ESR 153.6, Thunderbird 159, and Thunderbird 153.6.
-
CVE-2026-117970
Race condition in the Disability Access APIs component. This vulnerability was fixed in Firefox 159, Firefox ESR 153.6, Thunderbird 159, and Thunderbird 153.6.
-
CVE-2026-115479
baserCMS is a website development framework. In versions 5.2.5 before 5.3.2 and 5.4.0 before 5.4.2, an unauthenticated visitor can supply a comma-separated address list in a public mail form's email field and cause a single auto-reply to be addressed to multiple third parties. MailMessageMailer::sendFormToUser() passes the visitor-controlled userMail value to parseAndNormalizeMailAddresses(), which splits it with explode(',') and supplies every resulting address to setTo(). Newly created forms with an email-type field are affected when the separate, optional VALID_EMAIL server-side validation is not enabled; browser type=email validation does not prevent crafted requests. The bundled contact form enables VALID_EMAIL and is not affected. The auto-reply uses the site's sender identity and includes partially attacker-controlled content, allowing abuse of the site to deliver unsolicited mail. This issue has been patched in versions 5.3.2 and 5.4.2.
-
CVE-2026-126189
Soup Sieve is a CSS selector library designed to be used with Beautiful Soup 4. Prior to 4.0.1, Soup Sieve repeatedly searches descendant subtrees for each candidate evaluated by :has() and repeatedly reevaluates a parent containing :has(> ...) for each child, causing quadratic matching on deeply nested or wide attacker-influenced HTML or XML. Applications using ordinary fixed selectors through select(), select_one(), or Beautiful Soup CSS APIs against untrusted documents can have a worker occupied for seconds or minutes, while applications that process only trusted documents are not affected. Selectors compiled with NOCACHE continue to use the old uncached behavior. This issue is fixed in version 4.0.1.
-
CVE-2026-62337
SNARK Verifier is a generic (S)NARK verifier. Prior to 0.2.5, the Halo 2 aggregation circuit can mishandle proof points encoded as the point at infinity. `EcPointLoader::multi_scalar_multiplication` places a `Value::Assigned(_)` base paired with `Value::Constant(ONE)` into its unscaled variable-base list and passes it directly to `sum_with_const`. The default shim delegates to halo2-lib `EccChip::sum`, whose incomplete affine addition assumes nonidentity input points. A prover-controlled proof point encoded as `(0, 0)` can therefore enter the final aggregation sum through the scalar-one shortcut and affect the MSM or accumulator computation. Practical exploitability depends on the surrounding aggregation and final-decider checks; the advisory does not establish universal acceptance of invalid proofs. This issue is fixed in version 0.2.5.
-
CVE-2026-118764
minimatch is a minimal matching utility for converting glob expressions into JavaScript RegExp objects. Starting in 7.0.0 and prior to 11.0.1, src/index.ts implements firstPhasePreProcess() so that, at optimizationLevel 2 or higher, it repeatedly expands attacker-controlled **/.. pattern segments inside a fixpoint loop, allowing generated globParts entries to be processed and expanded again. secondPhasePreProcess() then compares the expanded entries pairwise, so a short malicious glob can cause excessive CPU and memory consumption during pattern compilation. This issue is fixed in version 11.0.1.
-
CVE-2026-107145
The twofactor_gauthenticator plugin for Roundcube is vulnerable to brute-forcing of the second-factor code. Due to the lack of rate limiting, an attacker who obtains user's password can submit arbitrary 6-digit codes, resulting in second-factor bypass and unauthorized access to the system. This issue was fixed in commit 4640889.
-
CVE-2026-116605
A vulnerability was found in SourceCodester Online Food Ordering System 1.0. The impacted element is an unknown function of the file order.php. The manipulation of the argument price/title results in client-side enforcement of server-side security. It is possible to launch the attack remotely. The exploit has been made public and could be used.