Vulnerabilities
All
My stack
Known exploited
Public exploit
Critical (9+)
Likely exploited (EPSS ≥ 50%)
Changed, 24 h
New, 24 h
Sort:
risk
modified
published
CVSS
EPSS
| CVE | CVSS | EPSS | Vendor, product | Description | Modified |
|---|---|---|---|---|---|
| CVE-2026-93834exploit | 8.8 high | — | red hat, red hat enterprise linux 10, red hat enterprise linux 6, red hat enterprise linux 7 | A use-after-free vulnerability was found in QEMU's 9pfs subsystem. A race condition between the main thread and a worker thread when processing concurrent Tlcreate and Twalk requests allows a malicious guest user to craft a fid path containing stale heap data, bypassing directory traversal restrictions and escaping the shared directory boundary. This can lead to arbitrary host file read/write and code execution (VM escape) as the QEMU process user. | 12 h ago |
| CVE-2026-69694 | 7.0 high | 3.0% | microsoft, windows 10 version 1607, windows 10 version 1809, windows 10 version 21h2 | Deserialization of untrusted data in Windows IP Address Management (IPAM) Service allows an authorized attacker to elevate privileges locally. | 4 h ago |
| CVE-2026-77484 | 8.8 high | 2.0% | microsoft, microsoft sql server 2019 (cu 32), microsoft sql server 2019 (gdr), microsoft sql server 2022 (cu 26) | Deserialization of untrusted data in SQL Server allows an authorized attacker to execute code over a network. | 4 h ago |
| CVE-2026-65772 | 8.8 high | 2.0% | microsoft, microsoft dynamics 365 (on-premises) version 9.1 | Deserialization of untrusted data in Microsoft Dynamics 365 allows an authorized attacker to execute code over a network. | 4 h ago |
| CVE-2026-81385 | 8.8 high | 1.0% | microsoft, microsoft 365 apps for enterprise, microsoft office 2019, microsoft office ltsc 2021 | Deserialization of untrusted data in Microsoft Office Publisher allows an unauthorized attacker to execute code over a network. | 4 h ago |
| CVE-2026-47297 | 8.1 high | 1.0% | microsoft, microsoft sql server 2019 (cu 32), microsoft sql server 2019 (gdr), microsoft sql server 2022 (cu 26) | Deserialization of untrusted data in SQL Server allows an unauthorized attacker to execute code over a network. | 4 h ago |
| CVE-2026-65812 | 6.8 medium | 1.0% | microsoft, microsoft teams for android | Insertion of sensitive information into sent data in Microsoft Teams for Android allows an authorized attacker to disclose information over a network. | 4 h ago |
| CVE-2026-69304 | 5.9 medium | 1.0% | microsoft, .net 10.0 runtime, .net 8.0 runtime, .net 9.0 runtime | Improper handling of highly compressed data (data amplification) in ASP.NET Core allows an unauthorized attacker to deny service over a network. | 4 h ago |
| CVE-2026-62874 | 10.0 critical | 0.00% | microsoft, azure billing | Insufficient verification of data authenticity in Azure Billing allows an unauthorized attacker to elevate privileges over a network. | 4 h ago |
| CVE-2026-118235 | 9.8 critical | — | wpazleen, simpleform – form maker to chat apps with quiz, poll and multi lead notifications | The SimpleForm – Form Maker to Chat Apps with Quiz, Poll and Multi Lead Notifications plugin for WordPress is vulnerable to Arbitrary File Upload in all versions up to, and including, 3.10.2 via the handle_upload function. This is due to missing server-side file key allowlist and pre-sanitization extension derivation that allows a trailing-dot filename to bypass all extension and MIME validation in handle_upload. This makes it possible for unauthenticated attackers to upload files that may be executable, which makes remote code execution possible. The only access control is a nonce that is printed into every public form's data-nonce attribute, making it trivially obtainable by any unauthenticated visitor on a page embedding the form shortcode. | 12 h ago |
| CVE-2026-63377 | 9.8 critical | — | fctr-id, okta-ai-agent | TAKO AI Agent for Okta is a local-first AI agent that helps Okta administrators query tenant data using natural language. Prior to 3.0.1-beta, `POST /api/auth/setup` in `src/api/routers/auth.py` allows an unauthenticated network attacker to create the first administrator account before the legitimate owner completes initial setup. The `initial_setup` function checks only the `is_setup` completion flag and requires no authentication, one-time bootstrap token, or network restriction. The default `docker-compose.yml` publishes port 8001 on all interfaces, the `Dockerfile` starts the listener on 0.0.0.0, and `src/api/main.py` mounts the authentication router without a global authentication guard. The endpoint invokes `create_auth_user` with `UserRole.ADMIN` and `set_auth_cookie`, issuing a valid `fctr_session` JWT cookie. Localhost-only CORS rules do not restrict non-browser HTTP clients, and the setup operation does not require access to a real Okta tenant. The attacker becomes the sole administrator and prevents legitimate first-run setup, can read Okta-synced user data, API tokens, chat history, and application configuration, modify users and Okta integration credentials, and lock out administrators or disable the service. Already completed setup is rejected. This issue is fixed in version 3.0.1-beta. | 22 h ago |
| CVE-2026-61620 | 9.8 critical | — | seancorfield, honeysql | HoneySQL turns Clojure data structures into SQL. From 2.0.0 until 2.7.1392, applications that pass user input or other arbitrary data into the HoneySQL `DSL` by calling `keyword` or `symbol` on that input can produce SQL strings that facilitate SQL injection when executed unchanged by a `JDBC` driver. Some formatters do not apply the existing suspicious-character check to the resulting entity names. The impact depends on the privileges of the database connection executing the generated statement. Input validation before constructing `keyword` or `symbol` entities is a workaround. The maintainer-designated fixed version is 2.7.1392. | 22 h ago |
| CVE-2026-116241 | 9.4 critical | — | pgadmin.org, pgadmin 4 | pgAdmin 4's Import/Export Data tool wraps a user's query export in a psql \copy (...) TO command and checks that the query's parentheses are balanced so that nothing can escape the wrapper. The check rejected a backslash inside a single-quoted string, but not inside a double-quoted identifier, where it treated the next double quote as closing the identifier. psql's \copy parser applies the connection's escape character to whichever quote opened a token, so with standard_conforming_strings off a backslash-escaped double quote does not close the identifier. pgAdmin's and psql's view of the quoting then diverged, a closing parenthesis hidden from the check ended the \copy (...) wrapper, and a following TO PROGRAM clause was executed by psql on the pgAdmin host, as the operating-system account running pgAdmin, before the server rejected the statement. An ordinary database role can turn standard_conforming_strings off for itself, and the Import/Export Data permission is granted to non-administrators by default. This is a bypass of the fix for CVE-2026-17566. The fix rejects any backslash inside a double-quoted identifier, as was already done for single-quoted strings, since its meaning depends on a server setting that cannot be known reliably when the query is validated. This issue affects pgAdmin 4: before 9.19. | Fri 30 Oct |
| CVE-2026-116557 | 9.1 critical | — | murgroland, wp booking system – booking calendar | The WP Booking System - Booking Calendar plugin for WordPress is vulnerable to unauthorized modification of data due to a missing capability check on the wpbs_action_ajax_calendar_remove_note() function in versions up to, and including, 2.1.0.5 This makes it possible for authenticated attackers, with Subscriber-level access and above, to delete arbitrary calendar notes belonging to any calendar. | 13 h ago |
| CVE-2026-62849 | 9.1 critical | — | mqttsnet, thinglinks | ThingLinks is an enterprise IoT platform. Prior to 1.4.0, an unauthenticated SQL injection vulnerability in the `TDengine TDS APIs` allows request-controlled JSON field values submitted to `/anyUser/tds/insertTableData` to reach unsafe SQL construction in the `TDS mapper`. A remote unauthenticated attacker can tamper with TDengine time-series data, including inserting unintended rows or overwriting existing readings. The affected surface is the TDS module's insert-table-data flow and TDengine mapper SQL generation. This issue is fixed in version 1.4.0. | 22 h ago |
| CVE-2026-55598 | 9.1 critical | — | madeofpendletonwool, pinepods | PinePods is a complete, self-hosted podcast management system written in Rust. Prior to 0.9.0, an authenticated user can subscribe to an attacker-controlled RSS feed through `POST /api/data/add_podcast` in `rust-api/src/handlers/podcasts.rs`. `get_podcast_values` and `add_episodes` in `rust-api/src/database.rs` parse the feed with `feed-rs` and store the `<enclosure url>` verbatim in `Episodes.episodeurl` without destination host/IP validation. `POST /api/data/download_podcast` invokes `spawn_download_podcast_episode` and `download_episode_and_wait()` in `rust-api/src/services/tasks.rs`, which sends a `reqwest` GET to the stored URL and saves and records the response in `DownloadedEpisodes` for later retrieval. The `estimate_duration_from_audio_url_async()` path in `rust-api/src/database.rs` also sends an unvalidated HEAD during feed add/refresh. Default redirect following permits an initially external enclosure URL to redirect to an internal target. A low-privilege user can make server-side GET/HEAD requests to reachable loopback, private-network and cloud metadata services and read downloaded internal responses, enabling internal-service enumeration, confidential-data disclosure and possible metadata credential theft. Episode downloading requires server downloads to be enabled, which is the default; self-service signup is supported. This issue is fixed in version 0.9.0. | Fri 30 Oct |
| CVE-2026-76010 | 8.8 high | — | smub, photo gallery, sliders, proofing and themes – nextgen gallery | The Photo Gallery, Sliders, Proofing and Themes – NextGEN Gallery plugin for WordPress is vulnerable to Remote Code Execution in all versions up to, and including, 4.4.1 via the render_image function. This is due to unrestricted mass assignment of caller-supplied meta_data keys onto the image model combined with missing realpath containment in path construction and a bypassable signature check on the public image delivery route. This makes it possible for authenticated attackers, with custom-level access and above, to execute code on the server. Exploitation requires the authenticated attacker to own at least one NextGEN gallery — a condition that can be satisfied by any Subscriber who was ever granted the 'NextGEN Manage gallery' capability, and that access is retained indefinitely due to the stored gallery-author equality fallback in _can_manage_gallery() even after the capability is revoked. | 11 h ago |
| CVE-2026-116086 | 8.8 high | — | no CPE yet | A flaw was found in konflux-ci. An attacker with permission to trigger automated workflow runs can cause an information disclosure by providing crafted issue titles to workflow scripts. The affected script incorporates user-provided titles directly into data queries without adequate sanitization, allowing the attacker to inject malicious filter commands. As a result, sensitive information such as API credentials, deployment keys, and internal notes returned from repository queries may be exposed in workflow logs. | 11 h ago |
| CVE-2026-63378 | 8.8 high | — | fctr-id, okta-ai-agent | TAKO AI Agent for Okta is a local-first AI agent that helps Okta administrators query tenant data using natural language. Prior to 3.0.1-beta, any authenticated user with a valid `fctr_session` cookie can submit arbitrary Python source in the `script_code` field of `POST /api/react/execute-script`, without administrator privilege. In `src/api/routers/react_stream.py`, the source is stored with `skip_discovery=True`, bypassing AI-based content filtering, and is passed to `validate_generated_code` in `src/utils/security_config.py`. `BLOCKED_PATTERNS` recognizes direct `__import__` calls but misses the import name passed as a string, `getattr` is in `ALLOWED_BUILTINS`, and the AST checker handles `ast.Name` and `ast.Attribute` targets but fails to handle `ast.Call` targets. Dynamic `getattr` call chains through `__builtins__` therefore evade validation. The `_write_temp_script` helper writes the accepted code to disk, and `asyncio.create_subprocess_exec()` executes it as the application process user, which is root in the default container. Command output is returned as Server-Sent Events through `GET /api/react/stream-react-updates?process_id=<id>`. The resulting container compromise can disclose environment secrets including Okta API tokens and the JWT key, expose the SQLite database, and permit backend-service pivoting and persistence. This issue is fixed in version 3.0.1-beta. | 22 h ago |
| CVE-2026-63396 | 8.7 high | — | roostorg, coop | Coop is an open source review and moderation tool from ROOST. Prior to 1.0.0, an unauthenticated actor can query `Query.allOrgs` through `/api/v1/graphql` on a publicly accessible deployment and obtain data for all organizations, including organization IDs, names, email addresses, website URLs and on-call alert email addresses. The `usersWhoCanReviewEveryQueue` field also exposes reviewing users' IDs, email addresses, first and last names, roles, permissions and creation dates. Running the service on an authenticated private network prevents outside actors from directly querying the GraphQL API. This issue is fixed in version 1.0.0. | 22 h ago |
| CVE-2026-63393 | 8.7 high | — | opensearch-project, sql | OpenSearch SQL supports querying OpenSearch data using SQL and Piped Processing Language. In the 2.x line from 2.8.0.0 until 2.19.6.0, and the 3.x line prior to 3.7.0.0, a remote authenticated user with basic read/search permissions can provide a crafted `cursor` parameter to the `plugins/sql` endpoint and trigger unsafe deserialization of untrusted data, resulting in arbitrary code execution on the server. This issue is fixed in repository releases 2.19.6.0 and 3.7.0.0. | 22 h ago |
| CVE-2026-62990 | 8.7 high | — | shopperlabs, shopper | Shopper is a Headless e-commerce Admin Panel. In versions from 2.0.0 prior to 2.10.1 and in the 3.0.0 prereleases beta.1 and beta.2, authenticated back-office staff with a content-upload permission can store executable media on the administrator's own origin. In the 2.x implementation, relevant upload permissions include `edit_products` or `add_brands`, and `packages/admin/config/media.php` allows `image/svg+xml` in every media collection, including image fields whose image/* validation also accepts SVG. The product files collection in `packages/admin/src/Livewire/Components/Products/Form/Files.php` has no MIME restriction and accepts `text/html` as well. Media is served inline from the `public` disk under `/storage/` with its own content type, without `Content-Disposition: attachment` or a `Content-Security-Policy`. When an administrator opens an uploaded HTML or SVG file, stored script executes in that administrator's session. In 2.x, the script can invoke `packages/admin/src/Livewire/SlideOvers/CreateTeamMember.php`, whose `store()` checks `access_setting` held by the viewing administrator but trusts client-supplied `role_id`, to create a new attacker-controlled administrator with a chosen password. This account persists after the uploaded file is deleted and gives access to customer data, orders, settings, and user/role management; the same script can exfiltrate data or grant permissions to the attacker's account. Direct account creation by the low-privileged uploader is blocked, so exploitation requires the administrator to open the file. The 3.x prerelease line already excludes SVG but still allows executable product files. This issue is fixed in versions 2.10.1 and 3.0.0-beta.3. | 23 h ago |
| CVE-2026-62284 | 8.6 high | — | osac-project, osac-operator | OSAC operator is a Kubernetes operator. In the affected revision identified as 8220502, the `console-proxy` `RemoteConfigResolver` parses Kubernetes Secret kubeconfig data with `clientcmd.RESTConfigFromKubeConfig` without sanitizing the resulting `rest.Config`. An attacker who can create or modify a Secret labeled `osac.openshift.io/remote-cluster-kubeconfig`: "true" in a namespace where the proxy's RoleBinding grants Secret read access can supply `users[].user.exec`. When `rest.TransportFor` builds the HTTP transport for an upstream WebSocket connection, client-go invokes that exec credential plugin, spawning an arbitrary process as the osac-`console-proxy` service account inside the pod. Attacker-controlled `clusters[].server` also enables server-side request forgery from the proxy pod. Remote and auto VM cluster modes are affected, with auto the default; exclusively `local` mode does not read kubeconfig Secrets and is unaffected. The credential-plugin execution path is patched in commit 7e4222b; that patch does not address the described server-address SSRF path. | 22 h ago |
| CVE-2026-102212 | 8.4 high | — | opengear, im7200, cm7100, acm7000 | Heap-based buffer overflow in Opengear IM72XX, CM71XX, and ACM70XX firmware versions 4.8.0 through 5.5.2 on console server appliances allows an unauthenticated remote attacker to crash the index.cgi process via a crafted multipart/form-data POST request containing an overlong Content-Disposition name field. | 4 h ago |
| CVE-2026-63182 | 8.1 high | — | litesaml, lightsaml | LightSAML implements SAML 2.0 data model classes, XML serialization and deserialization, XML security, certificate support, and message bindings. Prior to 4.7.1 and 5.0.1, LightSAML and xmlseclibs disagree about which assertion an inbound SAML signature covers. `Response::deserialize` uses `manyElementsFromXml` to collect only direct-child `saml:Assertion` elements of `samlp:Response`, and `AssertionSignatureValidatorAction` validates their `ds:Signature` using an out-of-band trusted IdP credential resolved by Issuer, not a key from the document. `XMLSecurityDSig::validateReference()` removes the enveloped signature from the shared DOM, but `processRefNode()` resolves `Reference URI="#id"` document-wide through an XPath of the form `//*[@Id]` and takes `item(0)`, which need not be the assertion whose claims LightSAML consumes. An attacker who has captured one genuinely signed assertion, such as an earlier login response, can place its signature-stripped bytes earlier in `samlp:Extensions` and present a forged direct-child assertion with the same ID and copied genuine signature. Validation then covers the hidden genuine bytes while LightSAML surfaces unsigned attacker-controlled `Subject/NameID`, `Conditions`, and `Attributes`, enabling authentication bypass or privilege escalation even when signature validation is required. The behavior was confirmed in versions 5.0.0 and 4.6.1. This issue is fixed in versions 4.7.1 and 5.0.1. | 22 h ago |
| CVE-2026-73014 | 7.8 high | 0.00% | microsoft, windows 10 version 1607, windows 10 version 1809, windows 10 version 21h2 | Missing authorization in Data Sharing Service Client allows an authorized attacker to elevate privileges locally. | 4 h ago |
| CVE-2026-56198 | 7.8 high | 0.00% | microsoft, windows 11 version 24h2, windows 11 version 25h2, windows 11 version 26h1 | Out-of-bounds read in Microsoft Trace Data Helper allows an authorized attacker to elevate privileges locally. | 4 h ago |
| CVE-2026-63414 | 7.6 high | — | shopify, shopify_app | Shopify App is a Rails engine for building Shopify applications. From 22.1.0 until 23.0.3, `token-exchange` authenticated controllers can expose a request-derived `shop` context instead of the identity from a verified Shopify ID token or active Shopify session. `current_shopify_domain` can resolve to the sanitized `shop` query parameter, so a request authenticated for one `shop` can cause application code to treat a different requested `shop` as current. Affected applications use `shopify_app` `token-exchange` authentication, accept authenticated requests with a `shop` query parameter, and rely on `current_shopify_domain` or `params[:shop]` for authorization, tenant lookup, stored access-token selection or `shop`-scoped data access. An attacker with a valid Shopify session token for one `shop` can supply another `shop` in that parameter and potentially obtain unauthorized application data or perform `shop`-scoped actions, depending on the application's authorization model. This issue is fixed in version 23.0.3. | 21 h ago |
| CVE-2026-62297 | 7.5 high | — | agamnentzar, ag-psd | ag-psd is a JavaScript library for reading and writing Photoshop Document files. Before the allocation-limit remediation attempt in version 31.0.0, `readPsd()` commits image and layer buffers from attacker-declared PSD geometry before validating the declarations against the actual available data or sane allocation limits. The `psdReader.js` implementation sizes off-heap typed arrays using `width`, `height`, `channels`, `depth` and `per-layer rectangles`, so a tiny PSD declaring enormous geometry can force multi-gigabyte memory allocation before a later bounds error. The reported 52-byte input on Node 24 produced approximately 3.17 GB peak RSS before a RangeError; under memory pressure the process can be terminated before the error can be caught, and concurrent inputs can exhaust memory. Services reading untrusted PSDs are exposed to denial of service. This differs from genuinely large PSD files whose memory demand is proportional to real input content. Version 31.0.0 adds a composite-image allocation limit, but complete coverage of the separately described layer-rectangle allocation path is not established and requires curator residual/scope review. | 23 h ago |
| CVE-2026-62966 | 7.5 high | — | esp32async, espasyncwebserver | ESPAsyncWebServer is an asynchronous HTTP and WebSocket server library for ESP32, ESP8266, RP2040 and RP2350. Prior to version 3.11.2, the multipart parser in `src/WebRequest.cpp` frees `_itemBuffer` and sets it to NULL when `BOUNDARY_OR_DATA` matches the closing boundary of a file part, before confirming the following terminator, while leaving `_itemIsFile` true. If the next byte is neither the final closing dashes nor a clean CRLF, the `DASH3_OR_RETURN2` or `EXPECT_FEED2` rewind branch re-injects provisionally consumed bytes through `itemWriteByte`. Because `_itemIsFile` remains true, `_itemBuffer`[`_itemBufferIndex`++] writes through a null pointer. On ESP32 this causes a `StoreProhibited` exception, panic and device reboot, producing remote denial of service. A route with a non-trivial body/upload handler, such as a file-upload or web-OTA endpoint, must be present. This issue is fixed in version 3.11.2. | 23 h ago |
| CVE-2026-116250 | 7.3 high | — | pgadmin.org, pgadmin 4 | In server mode pgAdmin 4 validates a path a user gives for a file in their storage directory with check_access_permission() and then opened it by name. Several places opened it with a plain open() or handed the path to a utility that opened it later: the servers.json export and import and the import file listing, and the files given to pg_dump, pg_dumpall, pg_restore and psql by Backup, Restore and Import/Export Data. A symbolic link swapped in at the file or at any parent directory between the check and the open redirected the operation outside the storage directory, so a file could be written, overwritten or read, or loaded into a table, as the operating-system account running pgAdmin. A hard link to an outside file, which O_NOFOLLOW does not detect, and a FIFO, which could hang a request, had the same effect. CVE-2026-86861 fixed this for the File Manager's save_file; these were the remaining paths. Exploitation requires creating such a link or FIFO inside the user's storage directory. pgAdmin offers no way to do so, and the per-user storage directories belong to the pgAdmin account, so this requires write access to the storage backend as a different operating-system user, for example a shared or network-mounted storage directory, and winning the race. The fix opens each file by walking down from the storage directory one component at a time with O_NOFOLLOW, accepting only a regular file with one link, and passes the utilities the open descriptor (/dev/fd/N) on Linux. Error messages no longer include the server path. This issue affects pgAdmin 4: before 9.19. | Fri 30 Oct |
| CVE-2026-63224 | 7.2 high | — | openremote | OpenRemote is an open-source internet-of-things platform. Prior to 1.26.0, the `datapoint export` API constructs `CSV crosstab` SQL by concatenating an `asset name` and `attribute name` into a `COPY` query that uses `crosstab` and streams `TO STDOUT`. The `asset name` enters both a double-quoted column identifier and a category query enclosed by the fixed `$cat$` delimiter. Escaping single quotes in one string context does not escape embedded double quotes in identifiers or protect that dollar-quote boundary. An authenticated user permitted to create or rename an asset and read/export datapoints for one of its attributes can store SQL metacharacters in the name and request a `crosstab` export, causing injected SELECT output to be returned in the normal ZIP/CSV response. This enables database exfiltration without direct database or filesystem access or token forgery. In a multi-tenant deployment it can expose other tenants' data if the manager's database role can read shared tables; integrity and availability effects are not established by the demonstrated read primitive. This issue is fixed in version 1.26.0. | 22 h ago |
| CVE-2026-116252 | 7.1 high | — | pgadmin.org, pgadmin 4 | pgAdmin 4's materialized view refresh passed the target database's name to psql's --dbname argument unchanged. libpq treats a --dbname value containing "=" (or a postgresql:// URI) as a full connection string, whose keywords override the host, port and other settings pgAdmin had supplied. A database whose name is such a string, which any role with CREATEDB can create, therefore redirected the psql process pgAdmin starts for the refresh to a host and port of the creator's choosing, and the connecting user's saved server password was presented to that endpoint. It also let the pgAdmin host be made to open connections to arbitrary addresses. The same weakness was fixed for Backup, Restore, Maintenance and Import/Export Data in earlier releases (CVE-2026-86862); the materialized view refresh was the remaining caller. The fix routes the name through the same database_conninfo() helper, so libpq treats it strictly as a database name. This issue affects pgAdmin 4: before 9.19. | Fri 30 Oct |
| CVE-2026-116240 | 7.1 high | — | pgadmin.org, pgadmin 4 | In server mode pgAdmin 4 validates a path into the user's storage directory with check_access_permission(), which builds the absolute path with get_abs_path(). Several callers then acted on a path they built differently, so the path checked and the path used could name different files: - Opening a file in the Query Tool, the Google client secret and, on Windows, Import/Export Data, Backup, Restore and saving a file joined the path to the storage directory after stripping leading slashes and backslashes. On POSIX a path checked as a name inside storage could become an absolute path once stripped, so any file the pgAdmin account can read could be opened; on Windows a path naming a drive discarded the storage directory, so these could read, or write, outside it. - Listing a folder decoded the path a second time after the check, so a folder named like an encoded "..", which a user can create, listed any directory on the host. - Download, delete, rename and the file-exists check appended the path to the storage directory without a separator, so in a deployment where one user's storage directory name is a prefix of another's, another user's files could be read, deleted or moved; download also checked the path appended to itself. - Creating a folder could create one in another user's directory. The fix makes check_access_permission() return the path it checked and every caller act on that path, decodes folder paths once, and opens, deletes and renames files relative to directories opened from the storage directory without following symbolic links. Desktop mode is not affected. This issue affects pgAdmin 4: before 9.19. | Fri 30 Oct |
| CVE-2026-62965 | 7.0 high | — | sabnzbd | SABnzbd is an open source automated Usenet download tool. Prior to 5.1.0Beta3, orphan-job management in `sabnzbd/api.py` accepts the user-controlled `value` parameter without path confinement before `os.path.join()` with `download_dir`. An absolute path discards that base directory. An attacker with a valid API key can invoke `mode=status` with `name=delete_orphan` to reach `_api_delete_orphan` and `remove_all()` in `sabnzbd/filesystem.py`, which uses `shutil.rmtree()` for recursive deletion. The advisory also identifies `_api_add_orphan` and `_api_delete_all_orphan` as affected by the path-validation issue, while its concrete arbitrary-deletion trace is through `_api_delete_orphan`. The API key is stored in plaintext in `sabnzbd.ini`. The attacker can delete files or directories writable by the SABnzbd process, including configuration, queue data, and downloads, but system files outside the regular user's ownership are not affected. This issue is fixed in version 5.1.0Beta3. | 23 h ago |
| CVE-2026-55613 | 6.9 medium | — | openwrt, uhttpd | uhttpd is a small HTTP server. In unpatched repository snapshots, `uh_ubus_data_send()` in `ubus.c` can desynchronize HTTP request boundaries when a `POST /ubus` JSON body arrives in multiple TCP segments. An initial segment completing a JSON object sets `du->jsobj`; a later segment takes the existing-object parse-error path and returns zero bytes consumed. The caller neither drains the remaining declared `Content-Length` nor sets `connection_close`. An HTTP request embedded in the unread body suffix is consequently processed as a new request on the same keep-alive connection. The reported reproducer returned a JSON-RPC parse error followed by a second HTTP response. This application-layer body-handler flaw is distinct from Transfer-Encoding header case handling and chunk-length state reset, and affects deployments enabling ubus, including the default OpenWrt configuration and potential proxy/CDN deployments. Repository patches 7b1bec45826bd78c8afc993435bdc0f1df2fe399 and b2285de129b9adc8b12f84f9a27c16c21f2b57a5 address this issue. | 22 h ago |
| CVE-2026-62335 | 6.9 medium | — | openvm-org, openvm | OpenVM is a performant and modular zkVM framework built for customization and extensibility. Before 1.7.0 in the 1.x line and in affected 2.0 and 2.1 prereleases, `OpenVmHalo2Verifier.verify` accepts `appExeCommit` and `appVmCommit` as `bytes32` values without checking that either is less than the `BN254` scalar field modulus. Although public-value and proof-data lengths are checked, callers can provide non-canonical 256-bit encodings greater than or equal to that modulus at the Solidity wrapper boundary. Applications relying on the wrapper can therefore accept non-canonical app executable and app VM commitments, creating ambiguity between the supplied byte-level values and the field elements used by the verifier. This issue is fixed in versions 1.7.0, 2.0.0-rc.3 and 2.1.0-preview. | Fri 30 Oct |
| CVE-2026-118598 | 6.5 medium | — | gnome, epiphany | A logic error in Epiphany's site permission handling can cause camera and microphone access to be granted when the user only approves a different permission. In src/ephy-window.c, set_permission() stores the user's Allow/Deny choice after a WebKit permission dialog. Types CLIPBOARD and WEBSITE_DATA_ACCESS are not stored by ephy_permission_is_stored_by_permissions_manager(), so control fell into an else-if that was written as: } else if (EPHY_PERMISSION_TYPE_ACCESS_WEBCAM_AND_MICROPHONE) { Because that expression is a non-zero enum constant, the branch always ran for those types and wrote EPHY_PERMISSION_PERMIT for both ACCESS_WEBCAM and ACCESS_MICROPHONE for the site origin. The user only saw a Clipboard or Website Data Access dialog. A malicious site that obtains approval for clipboard or third-party storage access can afterwards use getUserMedia without a further camera/microphone prompt, exposing audiovisual confidentiality. Requirements: the victim must visit a malicious page and click Allow on a clipboard or website-data-access prompt. Affected: Epiphany >= 45.0 (bug present at least through 51.x). Fixed: https://gitlab.gnome.org/GNOME/epiphany/-/merge_requests/2249 (commit 480c3648). Upstream: https://gitlab.gnome.org/GNOME/epiphany/-/work_items/2976 Reporter acknowledgment: Fernando Munoz | 11 h ago |
| CVE-2026-62991 | 6.5 medium | — | shopperlabs, shopper | Shopper is a Headless e-commerce Admin Panel. Prior to 2.10.1, the customer detail route `/cpanel/customers/{user}/show` in `packages/admin/src/Livewire/Pages/Customers/Show.php` checks `read_customers` but looks up an arbitrary user ID without applying the `customers` role scope used by the listing. Authenticated staff holding `access_dashboard` and `read_customers` can change the user ID to retrieve administrator or staff profiles despite lacking `view_users`, which protects the separate team-management interface. The flaw exposes account profile data and permits user-ID enumeration through differing success, denial, and not-found responses. If the same role also holds `delete_customers`, `anonymizeAction()` operates on the unrestricted loaded user model and can enable destructive follow-up against non-customer accounts. This issue is fixed in version 2.10.1. | 23 h ago |
| CVE-2026-55098 | 6.5 medium | — | basekick-labs, arc | Arc is an open, SQL-native time-series database for telemetry. Prior to the repository-verified 26.06.2 repair, the remediation of `CVE-2026-47735` leaves a cross-database authorization bypass through DuckDB table-valued I/O functions. In `internal/database/sandbox.go`, `allowed_directories` uses `cfg.LocalStorageRoot`, configured by `storage.local_path`, while `enable_external_access=false` blocks access outside that entire multi-tenant root. `GetStoragePath` in `internal/storage/util.go` stores all databases beneath it. In `internal/api/query.go`, `extractTableReferences` skips names followed by an opening parenthesis and `checkQueryPermissions` allows queries with no extracted table references, including tokens with `permissions: []`. The `userSQLReadParquetPattern` denylist catches literal `read_parquet(` and the earlier protection also blocks `arc_partition_agg(`, but misses the `parquet_scan` alias and `glob`, `parquet_metadata`, `parquet_schema`, `read_blob`, `read_csv_auto`, `read_json` and `read_text`. A request to `POST /api/v1/query` using these unblocked functions passes `ValidateSQLRequest` and `convertSQLToStoragePaths` without tenant authorization, allowing reads and enumeration of other databases' time-series data, schema and per-column value-range statistics within the allowlisted root. The sandbox still blocks out-of-root OS files and HTTP SSRF, and the sibling SQLite token database outside the root is not exposed by this mechanism. The earlier 26.06.1 remediation does not comprehensively deny these I/O functions. Repository tag 26.06.2 contains the I/O-function repair and supplies the provisional remediation boundary; conflicting advisory affected-version statements require curator adjudication. | Fri 30 Oct |
| CVE-2026-119328 | 6.5 medium | — | immich-app, immich | Immich is a high-performance self-hosted photo and video management solution. Prior to 3.3.0, the unauthenticated machine-learning /predict endpoint in machine-learning/immich_ml/main.py accepted an options object and passed its values into model construction in machine-learning/immich_ml/models/base.py without restricting security-sensitive parameters. An attacker with network access to the service could set cache_dir to an arbitrary writable path, causing model downloads to write there and causing clear_cache() to pass that path to shutil.rmtree() after selected model-loading failures. Depending on filesystem permissions and the selected path, this could modify files in mounted data locations or recursively delete an attacker-selected directory. The model-name allowlist prevents this issue from becoming arbitrary-model deserialization or code execution. This issue is fixed in version 3.3.0. | Fri 30 Oct |
| CVE-2026-55147 | 6.3 medium | — | aces, loris | LORIS (Longitudinal Online Research and Imaging System) is a self-hosted web application that provides data- and project-management for neuroimaging research. In versions from 22.0.0 prior to 27.0.4 and from 28.0.0 prior to 28.0.2, the `API` for the `imaging_browser` module permits edits to imaging data without checking the frontend permissions that restrict editing in Imaging Browser. An authenticated `API` caller can therefore modify imaging data without the required edit authorization. Deployments that no longer use the `API` can disable it to remove access to this endpoint. This issue is fixed in versions 27.0.4 and 28.0.2. | 23 h ago |
| CVE-2026-110023 | 5.9 medium | — | peterschulznl, wp data access – app builder for tables, forms, dashboards, charts & maps | The WP Data Access – App Builder for Tables, Forms, Dashboards, Charts & Maps plugin for WordPress is vulnerable to Sensitive Information Exposure in all versions up to, and including, 5.5.88 via the 'rel_tab' parameter. This makes it possible for unauthenticated attackers to extract every row and column of the underlying database table backing any anonymously published app, including WordPress user password hashes, email addresses, and password-reset keys from wp_users. Exploitation requires that an administrator has published at least one app with its REST API authorization set to 'anonymous', which is the standard configuration for publicly embedded apps. | 12 h ago |
| CVE-2026-63402 | 5.9 medium | — | i18next | i18next is an internationalization framework for JavaScript. Prior to version 26.3.4, `deepExtend` in `src/utils.js` checks property existence with the in operator and recurses into inherited built-ins reached through `Object.prototype`. An application must explicitly pass attacker-controlled resource data to `addResourceBundle(lng, ns, data, deep, overwrite)` with both `deep: true` and `overwrite: true`. Source keys such as `hasOwnProperty`, `toString` and `valueOf` can then select shared inherited built-in function objects, and a nested value can overwrite a property of those objects process-wide. For example, making `Object.prototype.hasOwnProperty.call` non-callable causes subsequent uses, including i18next internals, to throw TypeError, resulting in unrecoverable denial of service and limited integrity impact. The `__proto__`/`constructor` guard does not prevent this mechanism because the corruption target is an inherited built-in function rather than the prototype object itself. Default calls without deep merging or with overwrite falsy are unaffected, and no standard i18next backend or integration uses the vulnerable configuration. This issue is distinct from the setPath/key-walking prototype-pollution flaws in i18next-fs-backend (CVE-2026-48713) and i18next-http-middleware (CVE-2026-48714). This issue is fixed in version 26.3.4. | 22 h ago |
| CVE-2026-11396 | 5.3 medium | — | woobewoo, product filter for woocommerce by wbw | The Product Filter for WooCommerce by WBW plugin for WordPress is vulnerable to Missing Authorization / Information Disclosure in all versions up to, and including, 3.4.6 via the 'filterSettings[default_query]' parameter of the 'filtersFrontend' AJAX action. The handler is registered as 'wp_ajax_nopriv_filtersFrontend' and performs no capability or nonce check; raw POST data is JSON-decoded into $filterSettings and the attacker-controlled 'default_query' array is unconditionally array_merge()'d into the WP_Query arguments, overriding the plugin's safe defaults (most notably post_status, post_type, posts_per_page, and nopaging). This makes it possible for unauthenticated attackers to enumerate non-public WooCommerce products (draft, future, pending) — including their count and rendered listing markup — and to force unbounded queries that load all products in a single request. | 13 h ago |
| CVE-2026-63401 | 5.3 medium | — | roostorg, coop | Coop is an open source review and moderation tool from ROOST. Prior to 1.0.2, `addAccessibleQueuesForUser` and `removeAccessibleQueuesForUser` accepted arbitrary `queueId` and `userId` values without organization scoping. An authenticated user belonging to one organization and holding MRT queue-edit permission could grant or revoke another organization's users' access to manual review queues. The existing queue-edit permission check did not bind the supplied identifiers to the caller's organization. The attacker could not read the other organization's queues or jobs, so this issue did not disclose confidential data. This issue is fixed in version 1.0.2. | 22 h ago |
| CVE-2026-63115 | 5.3 medium | — | tryghost, activitypub | ActivityPub is a multi-tenant ActivityPub server for Ghost, built with Fedify. From 1.0.0 until 1.2.6, the `Undo` handler allows a remote federated server to cause the service to act on a `follow` or `repost` as though the activity came from a different account. This permits removal of an existing `follow` relationship or `repost` belonging to another account without that account's involvement. No private data is exposed and content is not modified; the affected relationships can be re-established. Services proxying to Ghost's hosted ActivityPub server have already been updated, while self-hosted services require an updated image and redeployment. This issue is fixed in version 1.2.6. | 23 h ago |
| CVE-2026-62860 | 5.3 medium | — | zoeille, picsou-finance | Picsou is a self-hosted personal finance dashboard. In the affected login implementation, the `/api/auth/login` endpoint reveals whether a username exists through a timing discrepancy. In `backend/src/main/java/com/picsou/controller/AuthController.java`, `AuthController.login()` throws `BadCredentialsException` immediately for an unknown username, before `passwordEncoder.matches(...)`, whereas an existing username with an incorrect password incurs the `bcrypt` comparison cost. Both cases return the same generic `HTTP 401` response, but repeated latency measurements disclose valid usernames without credentials. Login rate limiting, a small user population, and deployments inaccessible from outside the local network limit practical exposure; the weakness does not itself read or modify account data, though enumerated usernames can assist credential stuffing or brute-force attacks. This issue is fixed in repository release 1.0.8. | 24 h ago |
| CVE-2026-117645 | 5.3 medium | — | phoenixframework, phoenix_html | Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') vulnerability in phoenixframework phoenix_html allows an attacker who controls attribute names to inject attributes and script into rendered HTML via Phoenix.HTML.attributes_escape/1. attributes_escape/1 in lib/phoenix_html.ex escapes attribute values but not attribute names. key_escape/1 passes a binary key through html_escape, which neutralizes quotes and angle brackets but not whitespace, so a key such as x onmouseover ends the attribute and starts a new one. An atom key is only dasherized, so any character passes, including quotes and >. Nested data, aria and phx keys take the same paths. An application is affected when it builds attribute keys from untrusted input and renders them through HEEx <tag {@attrs}>, content_tag/3, or a direct call. Before 3.0.0 the same code lived in Phoenix.HTML.Tag. This issue affects phoenix_html: from 1.0.0 before 3.3.5 and from 4.0.0 before 4.4.0. | 24 h ago |
| CVE-2026-11850 | 5.0 medium | — | red hat, red hat hardened images, red hat enterprise linux 10, red hat enterprise linux 6 | An integer underflow vulnerability was found in MIT krb5 in the berval2tl_data() function in plugins/kdb/ldap/libkdb_ldap/ldap_principal2.c. The function performs an unsigned subtraction (bv_len - 2) without a prior bounds check. When bv_len is 0 or 1, the subtraction wraps to a large value which is then truncated to uint16_t, yielding 0xFFFE (65534) or 0xFFFF (65535). The subsequent malloc succeeds and memcpy reads up to 65534 bytes from a 0-1 byte buffer, resulting in a heap out-of-bounds read. The attack vector involves a malicious or compromised LDAP KDB backend returning a krbExtraData attribute with bv_len < 2, triggering the underflow when the KDC or kadmind reads principal data. | 4 h ago |
50 of 55 · Next
Download these 55 as CSV · subscribe to this list.
EPSS scores: 2026-10-30. NVD cursor: 2026-10-29 00:22. The News page shows only notable CVEs (KEV, your stack, new criticals, EPSS over 50%, or active exploitation per SSVC); this table has all of them.