The internet is broken right now.
18 critical CVEs without a fix in the last 48 hours. The pager is warming up. 22 more already have a fix that someone still has to apply.
-
CVE-2026-134255
Admidio from 5.0.17 before 5.0.18 contains a stored cross-site scripting vulnerability that allows authenticated users to inject JavaScript by saving a crafted value in their STREET profile field. Attackers can break out of the unencoded title attribute in members_assignment_data.php with an onmouseover handler that executes in the session of role managers hovering over the address icon.
-
CVE-2026-134256
Admidio from 5.0.17 before 5.0.18 contains a missing authorization vulnerability that allows authenticated regular users to upload images into the announcements module by abusing the CKEditor upload handler. Attackers can send id=ann_description to ckeditor_upload_handler.php to store images under adm_my_files/announcements/images and retrieve them via show_image.php, abusing the application as an image host.
-
CVE-2026-134257
Admidio before 5.0.18 contains an insufficient session expiration vulnerability that allows attackers to keep account access after a forgot-password reset because User::setPassword() deletes sessions using the visitor ID instead of the reset user's usr_id. Attackers already holding a victim's session or remember-me cookie can continue using it after the victim resets the password, keeping authenticated access to the account.
-
CVE-2026-134258
Admidio 5.0.0 before 5.0.18 contains a path traversal vulnerability in the Documents & Files renameFile() action that allows authenticated users with folder upload rights to move files outside adm_my_files by supplying ../ sequences in adm_new_name. Attackers can relocate an uploaded SVG containing script to the web root, where it is served inline and executes stored cross-site scripting against administrators for account takeover.
-
CVE-2026-134259
Admidio from 5.0.16 before 5.0.18 contains a stored cross-site scripting vulnerability that allows users with folder upload rights to store JavaScript by uploading SVG files served inline. Attackers can upload a script-bearing SVG that downloadFile() serves as image/svg+xml inline without nosniff or CSP, executing on the Admidio origin for viewers including unauthenticated visitors.
-
CVE-2026-134247
SiYuan 3.8.0 before 3.8.10 contains an incorrect authorization vulnerability that allows anonymous attackers to read indexed asset text referenced only by publish-disabled attribute-view rows. Attackers can request /api/asset/getDocAssets on a public host document to obtain the leaked asset path, then query /api/search/getAssetContentByPath to retrieve the confidential file's full indexed text.
-
CVE-2026-134249
SiYuan before 3.8.10 contains an authentication bypass vulnerability that allows unauthenticated attackers to gain Administrator role by spoofing X-Forwarded-For through a loopback-trusted reverse proxy. Attackers can send requests with Host set to localhost and X-Forwarded-For set to 127.0.0.1 to read private note-history contents via the /history/*path route when no lock-screen password is set.
-
CVE-2026-134250
Grav Form plugin 1.1.0 before 9.1.40 contains a stored cross-site scripting vulnerability in Form::uploadFiles(), which sanitizes SVG only when a file's last extension is .svg and accepts markup under unrecognized extensions. Anonymous attackers can upload files like drawing.svg.foo through public any-type upload forms, running script in an administrator's session when opened, enabling super-admin creation on Grav 2.x with Admin2.
-
CVE-2026-134251
Grav API plugin from 1.0.16 before 1.1.11 contains an authorization bypass vulnerability that allows scoped API key holders to override page permission rules because checks read the owning account's authority instead of the key's scopes. Attackers can use keys minted on super-admin accounts to edit, delete or read pages that deny those actions, or use media-scoped keys to change media on read-only pages.
-
CVE-2026-134253
Grav API plugin from 1.0.0 before 1.1.4 contains a remote code execution vulnerability that allows publishers with media and page write rights to write theme template files through the blueprint-upload endpoint. Attackers can upload a file into a theme's Twig templates folder and include it from sandboxed page Twig to execute commands as the web-server user.
-
CVE-2026-134245
SiYuan 3.8.0 before 3.8.10 contains an incorrect authorization vulnerability in the serveAssets route that allows unauthenticated attackers to read password-protected document assets via name collisions. Attackers can request GET /assets/<name> when a visible document references the same relative asset name, retrieving the protected file's contents on public-publish instances with Basic Auth disabled.
-
CVE-2026-134246
SiYuan 3.6.0 before 3.6.2 and 3.7.3 before 3.8.10 contains a missing authorization vulnerability in the /api/block/getDOMText endpoint, which renders caller-supplied DOM without checking publish access for referenced image paths. When publish authentication is disabled, anonymous visitors can submit DOM referencing a known unpublished, non-encrypted asset path to retrieve its already-cached OCR text.
-
CVE-2026-133250
@fastify/middie is the middleware engine for the Fastify web framework. In versions up to and including 9.3.7, middie matches path-scoped middleware against its own reading of the request path, which can differ from the path Fastify's router uses to dispatch the request. When the two disagree, the router can route a request to a handler covered by a middleware mount while middie skips that middleware, so an unauthenticated client reaches a handler that an authentication, authorization, or rate-limiting middleware was meant to protect, with no special configuration required. The mismatch is triggered by empty parameter or wildcard values, request targets that do not begin with a slash, and literal percent characters in a mount path. The issue is fixed in @fastify/middie 9.3.8, and users should upgrade to 9.3.8 or later.
-
CVE-2026-133273
@fastify/swagger-ui is the plugin that serves Swagger UI documentation for Fastify applications. In versions 2.1.0 through 6.1.4, the authentication hooks that applications register to protect the documentation are applied to every route the plugin adds except the bundled static asset instance. When the option that serves bundled UI files from a local directory is enabled, that instance exposes every file in the directory without credentials, so a specification file an operator meant to keep behind the documentation login becomes reachable by an unauthenticated client who knows or guesses its name. Access is read-only and limited to files in that directory. The issue is fixed in @fastify/swagger-ui 6.1.5, which applies the same authentication hooks to the static asset route, and users should upgrade to 6.1.5 or later.
-
CVE-2026-133312
@fastify/rate-limit is the rate-limiting plugin for the Fastify web framework. Starting in version 11.1.0, the plugin documented a manual flow for limiting failed login attempts that peeks at the limiter, runs the password check, and consumes a slot only on failure. Because the peek and the consume are two separate store operations with the password check in between, concurrent requests can all pass the peek before any slot is consumed, so in versions 11.1.0 through 11.2.2 a burst of requests runs far more failed attempts than the configured limit, and the sequential path also allowed one attempt more than configured. The default per-request limiter is not affected. The issue is fixed in @fastify/rate-limit 11.3.0, which makes the manual flow atomic by reserving a slot in a single operation and refunding it only on success, and users should upgrade to 11.3.0 or later.
-
CVE-2026-133867
@fastify/express is a compatibility layer that lets Express middleware run inside the Fastify web framework. In versions 4.0.7 through 4.0.9, the plugin does not rewrite regular-expression mount paths to account for a parameterized plugin prefix, so when middleware is mounted with a regular expression inside a plugin registered under a parameterized prefix, the Express matcher treats the prefix parameter as literal text and never matches the real request path. As a result, Express silently skips that middleware while Fastify still runs the route handler, so an application that relies on regular-expression mounted middleware for authentication or authorization inside a parameterized-prefix plugin exposes protected routes to unauthenticated clients. No special configuration is required beyond a parameterized prefix and a regular-expression mount. The issue is fixed in @fastify/express 4.0.10, and users should upgrade to 4.0.10 or later.
-
CVE-2026-129850
@fastify/express is a compatibility layer that lets Express middleware run inside the Fastify web framework. In versions 4.0.0 through 4.0.9, when Fastify routing is configured to be case-insensitive, the router folds the request path to lower case in a way that maps certain non-ASCII Unicode characters to ASCII letters, but @fastify/express matches Express mount paths against the decoded path without applying the same folding. The two disagree, so a request that spells a letter of a mount path with an equivalent Unicode character is routed to the protected handler while the path-scoped middleware mounted on that path never matches and is skipped. An unauthenticated client can therefore reach a handler that an authentication, authorization, or rate-limiting middleware was meant to protect, though only when the application enables case-insensitive routing and the mount path contains an affected letter. The issue is fixed in @fastify/express 4.0.10, and users should upgrade to 4.0.10 or later.
-
CVE-2026-133120
@fastify/express is a compatibility layer that lets Express middleware run inside the Fastify web framework. In versions before 4.0.10, the plugin passes Express a request URL that does not always match the path Fastify's router dispatched, so path-scoped Express middleware can be skipped while the route handler still runs. Several request forms trigger the mismatch, including percent-decoding that rewrites a backslash to a slash or double-decodes an encoded percent sign, and a request target that begins with an asterisk, so the middleware and the handler disagree on the path or on parameter and query values. An unauthenticated client can therefore reach a handler that an authentication, authorization, or rate-limiting middleware was meant to protect, with no special configuration required. The issue is fixed in @fastify/express 4.0.10, and users should upgrade to 4.0.10 or later.
-
CVE-2026-124626
@fastify/express is a compatibility layer that lets Express middleware run inside the Fastify web framework. In versions 4.0.5 through 4.0.9, the plugin matches path-scoped Express middleware with a matcher that requires a non-empty value for a dynamic segment, while Fastify's router accepts an empty parameter or wildcard value. When a route and an Express middleware mount share a dynamic pattern, a request with an empty segment matches the Fastify route and runs the handler while the middleware mounted on that pattern is skipped, so on a scoped mount an unauthenticated request can reach a protected handler. No special configuration is required, and the practical impact depends on what the empty value selects in the application. The issue is fixed in @fastify/express 4.0.10, and users should upgrade to 4.0.10 or later.
-
CVE-2026-129504
@fastify/express is a compatibility layer that lets Express middleware run inside the Fastify web framework. In versions 4.0.0 through 4.0.9, the plugin joins a path-scoped middleware's mount path to the Fastify plugin prefix by string concatenation, so when the plugin prefix ends with a slash or the mount path omits its leading slash, the resulting Express mount path does not line up with the route the Fastify router serves. Express then fails to match and passes the request on while the Fastify router still dispatches the route, so the middleware is silently skipped. An application that uses path-scoped Express middleware for authentication or authorization inside a prefixed plugin can expose protected routes to unauthenticated clients, with no crafted request and no non-default option required, and string, array, and regular-expression mount paths are all affected. The issue is fixed in @fastify/express 4.0.10, and users should upgrade to 4.0.10 or later.
-
CVE-2026-133983
A vulnerability was detected in eigent-ai eigent up to 1.0.2. Affected by this vulnerability is an unknown functionality of the file backend/app/controller/chat_controller.py of the component Local Agent Control API. Performing a manipulation results in missing authentication. The attack may be initiated remotely. Upgrading to version 1.0.3 addresses this issue. The patch is named 59f43afcb1b0ba7b2de8c65d585124f196c4aa17. It is recommended to upgrade the affected component.
-
CVE-2026-124976
The Swift Performance Lite plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the '<attacker-chosen parameter name>' parameter in all versions up to, and including, 2.3.7.5 due to insufficient input sanitization and output escaping. This makes it possible for unauthenticated attackers to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. This requires that an administrator has configured at least one Cacheable AJAX Action via the plugin's AJAX Cache feature, a condition attackers can confirm by probing for a swift-performance: HIT response header.
-
CVE-2026-126705
The WP Job Portal – AI-Powered Recruitment System for Company or Job Board website plugin for WordPress is vulnerable to time-based SQL Injection via the 'columns[<attacker-chosen key>]' parameter in all versions up to, and including, 2.8.0 due to insufficient escaping on the user supplied parameter and lack of sufficient preparation on the existing SQL query. This makes it possible for authenticated attackers, with subscriber-level access and above, to append additional SQL queries into already existing queries that can be used to extract sensitive information from the database.
-
CVE-2026-124248
The GPTranslate – Multilingual AI Translation Agent for WordPress: Translate Your Site with AI plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'comment' parameter in all versions up to, and including, 2.34.17 due to insufficient input sanitization and output escaping. This makes it possible for unauthenticated attackers to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. Exploitation requires the plugin's non-default 'HTML DOM method' to be enabled for server-side translation, and a translation row for the target language-prefixed URL must exist to arm the output buffer that routes page content through the vulnerable Simple HTML DOM parser.
-
CVE-2026-130323
The Accept Stripe Payments plugin for WordPress is vulnerable to Stored DOM-Based Cross-Site Scripting via the 'href of any <a> element containing 'asp_action=show_pp&product_id='' parameter in all versions up to, and including, 2.1.4 due to insufficient input sanitization and output escaping. This makes it possible for unauthenticated attackers to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. Successful exploitation requires that a user clicks the attacker-placed comment link to trigger the sink; however, if the non-default 'Prefetch Payment Popup Scripts' option is enabled, execution occurs on page load with no user interaction required.
-
CVE-2026-120736
The AutomatorWP – No-Code Workflow Automation, Integration & Webhooks Plugin, now with AI plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'your-message (any CF7 form field used in a form_field: tag)' parameter in all versions up to, and including, 6.0.9 due to insufficient input sanitization and output escaping. This makes it possible for unauthenticated attackers to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. This requires an administrator to have configured a public 'Guest submits a form' automation whose Send-an-email action embeds a form_field tag, after which the payload fires the next time any user opens the affected automation's email log entry.
-
CVE-2026-120745
The Paid Membership Plugin, Ecommerce, User Registration Form, Login Form, User Profile & Restrict Content – ProfilePress plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'reg_bio' parameter in all versions up to, and including, 4.18.0 due to insufficient input sanitization and output escaping. This makes it possible for unauthenticated attackers to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. The reg_bio parameter is accepted by the registration handler's allowlist even when the rendered registration form does not include a biography field, meaning an unauthenticated attacker can supply the payload via a raw POST request without needing a visible biography input on the form.
-
CVE-2026-122228
The ProfilePress – Paid Membership, Online Courses (LMS), Ecommerce, User Registration & Login Form, User Profile & Restrict Content plugin for WordPress is vulnerable to Sensitive Information Exposure in all versions up to, and including, 4.18.0 via the 'sortby<form_id>, filter<form_id>' parameter. This makes it possible for unauthenticated attackers to extract reconstruct any listed member's full e-mail address one character at a time via a boolean LIKE oracle, and enumerate users holding arbitrary usermeta keys such as ppress_billing_phone. The e-mail address oracle is reachable on any default install that used the plugin's 'Create pages' onboarding notice, because pp_email_address is enabled as a search field by default and the resulting directory page is published with no capability or nonce gate.
-
CVE-2026-122264
The wpForo Forum plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'member[about]' parameter in all versions up to, and including, 3.2.4 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with subscriber-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. Exploitation requires the attacking subscriber to have met the default antispam threshold of at least 3 forum posts before the account editing page becomes accessible.
-
CVE-2026-123830
The wpForo Forum plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'data[display_name]' parameter in all versions up to, and including, 3.2.4 due to insufficient input sanitization and output escaping. This makes it possible for unauthenticated attackers to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. This requires public user registration to be enabled (users_can_register=1) and a valid _wpfnonce fetched from the public sign-up page, but is otherwise fully unauthenticated.
-
CVE-2026-123970
The wpForo Forum plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'thread[body]' parameter in all versions up to, and including, 3.2.4 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with subscriber-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. Successful execution of the injected script is limited to viewers whose forum access profile denies attachment viewing (va=0 or caa=0), meaning exploitation depends on an administrator having configured a custom forum access profile that permits reading but denies attachment viewing — this is not the default shipped configuration.
-
CVE-2026-124216
The Orphans plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'comment' parameter in all versions up to, and including, 3.4.5 due to insufficient input sanitization and output escaping. This makes it possible for unauthenticated attackers to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. Exploitation requires a two-step comment chain: a benign first comment must be approved under a given name and email so that WordPress's comment_previously_approved mechanism auto-publishes the attacker's payload comment without moderation.
-
CVE-2026-107747
The WP Custom Admin Interface plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'first_name' parameter in all versions up to, and including, 7.43 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with subscriber-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page.
-
CVE-2026-107771
The Speed Booster Pack ⚡ PageSpeed Optimization Suite plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'author' parameter in all versions up to, and including, 4.5.9.1 due to insufficient input sanitization and output escaping. This makes it possible for unauthenticated attackers to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. Exploitation requires that the plugin's Lazy load media option is enabled and that the attacker's comment is approved by a site administrator or via auto-approval.
-
CVE-2026-116518
The ElasticPress plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'orderby' parameter in all versions up to, and including, 5.3.5 due to insufficient input sanitization and output escaping. This makes it possible for unauthenticated attackers to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. The injected payload is stored in the ep_query_log transient for up to 24 hours after a single unauthenticated frontend GET request, and only fires when an administrator subsequently opens the Status Report page and expands the Failed Queries panel.
-
CVE-2026-109196
The FV Player 8 plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'data.videos[].fv_wp_flowplayer_field_splash' parameter in all versions up to, and including, 8.1.9 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with subscriber-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. The wp_ajax_fv_player_db_save handler's ownership and capability check is conditional on the presence of the data.update field; omitting it bypasses authorization entirely, allowing a Subscriber-level attacker to overwrite any existing video record using an enumerable video ID — and the valid nonce required by the endpoint is issued to every logged-in visitor of any page hosting the publicly accessible [fvplayer_editor] shortcode.
-
CVE-2026-119157
The PublishPress Capabilities: User Role Access Control, Admin Area Permissions plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'page_elements' parameter in all versions up to, and including, 2.53.0 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with subscriber-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. Successful exploitation requires an administrator to open a Profile Features capture window and save at least one of the attacker-supplied rows on the Profile Features screen, which represents the ordinary documented use of the feature and leaves no visible indication to the administrator that malicious data was submitted.
-
CVE-2026-119177
The Real Cookie Banner: GDPR & ePrivacy Cookie Consent plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'comment' parameter in all versions up to, and including, 5.4.2 due to insufficient input sanitization and output escaping. This makes it possible for unauthenticated attackers to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. Exploitation requires that the cookie banner has been activated by the site administrator (the plugin's primary onboarding step) and that at least one prior comment from the same name and email identity has already been approved, after which WordPress core's comment_previously_approved default auto-approves subsequent comments without administrator action.
-
CVE-2026-107654
The One to one user Chat by WPGuppy plugin for WordPress is vulnerable to generic SQL Injection via the 'userId' parameter in all versions up to, and including, 1.1.6 due to insufficient escaping on the user supplied parameter and lack of sufficient preparation on the existing SQL query. This makes it possible for unauthenticated attackers to append additional SQL queries into already existing queries that can be used to extract sensitive information from the database. The plugin's permission callback is trivially bypassed because it validates a JWT that the plugin itself signs using an attacker-supplied value from the guppy_guest_account cookie, and the guest code path (userType=0) skips the get_userdata() check that would confirm the userId corresponds to a real WordPress user.
-
CVE-2026-129344
Uncaught Exception vulnerability in phoenixframework phoenix allows an unauthenticated attacker to cause a denial of service via a client-selected legacy serializer on binary channel broadcasts. When an application broadcasts {:binary, data} payloads on a channel topic, any client that connects without a vsn parameter, or with vsn=1.0.0, and joins that topic makes those broadcasts fail. Phoenix.Channel.Server.send/4 encodes each broadcast with every subscriber's own serializer inside the PubSub dispatch loop, calling fastlane!/1 with no rescue, and the negotiated Phoenix.Socket.V1.JSONSerializer raises Protocol.UndefinedError on a binary payload. Every binary broadcast on the topic then crashes the broadcasting process and skips every subscriber dispatched after the attacker. For broadcasts arriving from other nodes the loop runs in the PubSub adapter worker, whose repeated crashes restart the PubSub supervisor and empty the node's subscription registry while sockets stay connected. This issue affects phoenix: from 1.5.7 before 1.5.16, from 1.6.0-rc.0 before 1.6.18, from 1.7.0-rc.0 before 1.7.25, and from 1.8.0-rc.0 before 1.8.17.
-
CVE-2026-128361
Improper Input Validation vulnerability in phoenixframework phoenix (Phoenix.Presence) allows an attacker who can join a channel to cause a denial of service via a non-string presence key. track/3,4 accepts any term as the presence key, such as a JSON object taken from a join payload. The key is converted with to_string/1 only later, in Phoenix.Presence.group/1, which raises Protocol.UndefinedError on a map. group/1 runs whenever presences are listed and when each presence diff is broadcast, together with the application's fetch/2, which can likewise raise on an unexpected key. Once such a key is tracked, every process that lists presences on the topic crashes, typically each channel that joins it, and presence diffs for the topic are lost. From 1.7.0 the diff runs in a Task.Supervisor.async task linked to the tracker shard, and Phoenix.Presence.handle_info/2 has no clause for its failure. The shard then terminates, taking down every channel and LiveView process tracked on it and erasing presence state for all its topics. Repeated joins exhaust the restart intensity and take down the application supervisor. This issue affects phoenix: from 1.2.0 before 1.5.16, from 1.6.0-rc.0 before 1.6.18, from 1.7.0-rc.0 before 1.7.25, and from 1.8.0-rc.0 before 1.8.17.
-
CVE-2026-132989
In JetBrains TeamCity before 2026.2, 2026.1.4, 2025.11.8 stored XSS on the SSH Keys page was possible
-
CVE-2026-126082
Active Storage makes it simple to upload and reference files in cloud services and attach those files to Active Record objects. Prior to 8.1.4.1 and 8.0.5.2, a Rails application that processes attachments with ImageMagick may allow an unauthenticated attacker to write arbitrary files on the server and, when attachments are stored with the Disk service, to read arbitrary files readable by the application process. Those files typically include secret_key_base and credentials for external systems, which may in turn allow escalation to remote code execution or lateral movement to those systems. ImageMagick selects a coder from the content of the file it is given, and some coders act on files named inside the image rather than on the image itself. Active Storage constrained none of ImageMagick's coders, and the ImageMagick analyzer runs when a file is attached, so an attacker who can upload a crafted file may be able to invoke one. Attaching the upload is sufficient; no variant, preview, or analyze call is required. One specific attack chain has been reported, but other attack chains may exist. An application is affected when it uses ImageMagick for Active Storage image processing through config.active_storage.variant_processor = :mini_magick, runs ImageMagick under a security policy that does not already deny the involved coders, and allows uploads from untrusted users. This issue is fixed in versions 8.1.4.1 and 8.0.5.2.