Vulnerabilidades exploráveis hoje
418,896na visualização atual
Score único que combina CVSS, presença no KEV e EPSS. Cada CVE com sua própria ficha — timeline da publicação até a exploração ativa.
No catálogo KEV1,786
Novos KEV · 24H0
Exploit Today ≥ 701,715
Distribuição · última janela
- Crítica4,730
- Alta18,254
- Média18,061
- Baixa1,977
Filtros
Janela
Severidade
Filtros
CVECVSSEPSSKEVRExploitTítuloMod.
CVE-2026-133165——
——0MISP contains an authorization and scope flaw in the ObjectReference model's captureReference method. When a user saves an event (via REST API, sync, or feeds), any object reference included in the payload is looked up by its UUID across the entire MISP instance rather than being scoped to the event being saved. An authenticated user with permission to add or edit events can therefore include the UUID of an object reference belonging to a different event—including events in other organizations that the user cannot view—and cause the system to overwrite that foreign reference's relationship_type, comment, deleted flag, and timestamp with attacker-supplied values. The affected reference is neither deleted nor recreated in the attacker's event; its fields are silently rewritten in place. This constitutes a cross-tenant integrity violation: data integrity of object references in events the attacker has no read or write access to is compromised. The vulnerability is reachable through any event save path (REST add/edit, inter-instance sync, feed ingestion) and requires only a valid authenticated session with event modification privileges and knowledge or discovery of a target reference UUID.1dCVE-2026-99800——
——0In the Linux kernel, the following vulnerability has been resolved:
io_uring: free the task context after a grace period
A task other than the owner can free a request, io_ring_exit_work()
running the local task_work of a DEFER_TASKRUN ring for example, or
io_do_iopoll() reaping completions. io_put_task() then takes the else
branch:
percpu_counter_sub(&tctx->inflight, 1);
if (unlikely(atomic_read(&tctx->in_cancel)))
wake_up(&tctx->wait);
put_task_struct(tctx->task);
The first line is what lets the owner leave io_uring_cancel_generic(),
and on exec it frees the tctx right after. Nothing holds that off for
the remaining three lines, the task reference only covers a tctx freed
along with the task_struct. It used to work out because
io_uring_del_tctx_node() took the uring_lock that the freeing side holds
across io_put_task(), until that moved to tctx_lock.
Free the tctx with kfree_rcu(), and enter an RCU read section in
io_put_task() before the count is dropped.12hCVE-2026-99799——
——0In the Linux kernel, the following vulnerability has been resolved:
io_uring/sqpoll: don't wait for the thread to exit
io_sq_thread_stop() wakes the SQPOLL thread and waits for it to exit.
The thread belongs to the thread group that created the ring, so a
SIGSTOP or the cgroup freezer holds it in get_signal(), and the wake up
doesn't change that. io_ring_exit_work() then sleeps in that wait until
someone continues or kills the process. iou_exit has 64 slots, so 64
closed SQPOLL rings of a stopped process keep every other ring from
being freed, for everyone.
Don't wait. Set the stop bit, wake the thread, and let it exit when it
gets to run again. The sqd can now outlive its last ring, so give it a
second refcount with one reference for the rings and one for the thread,
and free it with the last one. The thread can't reach a freed ring,
rings leave ctx_list before dropping their reference, and the io-wq hash
is refcounted.12hCVE-2026-1238918.1 ALT—
——0SDDC Manager contains an SQL-injection vulnerability due to improper sanitization of user-supplied input.
Affected versions: 9.1.x.x (fixed in 9.1.1.0200); 9.0.x.x (fixed in 9.0.2.0100); 5.x (fixed in 5.2.5).15hCVE-2026-1331144.3 MÉD—
——0Caddy – WooCommerce Side Cart & Free Shipping Bar plugin for WordPress 3.0.0 before 3.2.1 contains a missing authorization vulnerability in the saved-items/add REST endpoint that exposes non-public product data. Authenticated attackers with Subscriber-level access can enumerate product_id values to read names, prices, images and stock status of private, draft, hidden and password-protected products.1dCVE-2026-99798——
——0In the Linux kernel, the following vulnerability has been resolved:
io_uring: defer foreign io-wq enqueues to task_work
An IOPOLL ring without DEFER_TASKRUN can be reaped by a task other than
the one that submitted the I/O. If a polled request is marked
REQ_F_REISSUE, io_free_batch_list() calls io_queue_iowq() with the
original tctx. Ordinary task exit can tear down tctx->io_wq after
draining only the tracked requests. The live request pins the task, but
not its io-wq, so the foreign task can race io_wq_put_and_exit() and
enqueue through a freed or NULL queue.
Forward foreign userspace enqueues to the owning task before accessing
tctx->io_wq. Use a dedicated task_work callback so reissues still go
straight to io-wq, and fail them if task_work is running in cancellation
mode. Preserve the existing PF_KTHREAD cancellation path. This also
covers the other callers of io_queue_iowq() without changing the
worker-owned enqueue path.12hCVE-2026-99797——
——0In the Linux kernel, the following vulnerability has been resolved:
io_uring/cancel: drop sqd->lock while SQPOLL waits
An exiting SQPOLL thread calls io_uring_cancel_generic() with sqd->lock
held, and keeps it across cond_resched() and the sleep until nothing is
in flight anymore. A zero-copy send notification can't be canceled and
stays pending as long as the peer doesn't read, so the lock is held for
as long as a user cares to. io_ring_exit_work() needs it to park the
thread, and so do IORING_REGISTER_IOWQ_AFF and IOWQ_MAX_WORKERS. With
64 such rings sharing the thread, every iou_exit slot sleeps on the lock
and no ring gets freed for anyone.
Drop the lock around cond_resched() and the sleep, like the main loop of
the thread does. Each walk of ctx_list is still under one hold, and
nothing is kept across the unlock. The thread drops its io-wq, frees its
tctx and clears sqd->thread under one hold after the loop, so whoever
holds the lock and finds sqd->thread set can still use them.
[axboe: adapted to the schedule_timeout() based loop]12hCVE-2026-1331136.5 MÉD—
——0Gogs through 0.14.4 contains an authorization bypass vulnerability that allows collaborators with write access to push to protected branches by merging pull requests. Because MergePullRequest skips branch protection checks and PullRequest.Merge omits hook environment variables, attackers can merge commits into branches restricted to allowlisted users.22hCVE-2026-879566.1 MÉD—
——0When Private data source connect (PDC) or a secure SOCKS proxy is used, the MQTT data source plugin does not apply TLS to broker connections for the tls, ssl, tcps, mqtts and mqtt+ssl schemes. The configured TLS settings are ignored, and the broker username and password are sent unencrypted on the hop between the proxy and the broker. The server certificate is not verified, and a configured client certificate is not sent.1dCVE-2026-99796——
——0In the Linux kernel, the following vulnerability has been resolved:
io_uring/poll: cancel all matching requests in one pass
With IORING_ASYNC_CANCEL_ALL, __io_async_cancel() calls io_try_cancel()
until nothing is found, and io_poll_cancel() cancels one request per
call, starting each lookup at the beginning of the hash. A cancelled
request stays hashed until its task work runs, so every lookup walks
past the ones already cancelled. Cancelling K of N pending polls visits
about K*N/2 requests in one io_uring_enter(), with uring_lock held.
A single CANCEL_FD|ALL on 80000 polls on one eventfd takes 17.06s,
0.02s with this change. With voluntary preemption the soft lockup
watchdog fires:
watchdog: BUG: soft lockup - CPU#1 stuck for 26s! [repro:100]
RIP: 0010:io_cancel_req_match+0x19/0xa0
io_poll_cancel+0x73/0x170
io_try_cancel+0x7b/0xf0
__io_async_cancel+0x41/0x110
Walk the hash once and cancel every match, as io_cancel_remove() does
for futex and waitid. Cancel by user_data only needs its own bucket.
__io_async_cancel() still loops until nothing is found, so a request
that io-wq parks in the poll hash meanwhile is still cancelled.
A cancel that matches one request now walks the whole hash twice instead
of stopping at the match.12hCVE-2026-890958.1 ALT—
——0The Shortcodes and extra features for Phlox theme plugin for WordPress is vulnerable to Local File Inclusion in all versions up to, and including, 2.17.22 via the 'extra_template_path' parameter parameter. This makes it possible for unauthenticated attackers to include and execute arbitrary .php files on the server, allowing the execution of any PHP code in those files. This can be used to bypass access controls, obtain sensitive data, or achieve code execution in cases where .php file types can be uploaded and included. The slug-based allowlist introduced as the fix for CVE-2023-37888 constrains only the template_part_file filename suffix and does not validate the extra_template_path directory component, meaning path traversal sequences are fully functional against this parameter.5hCVE-2026-1331178.2 ALT—
——0WebToffee Product Feed before 2.4.4 for WordPress contains a use of hard-coded cryptographic key vulnerability that allows unauthenticated attackers to bypass the URL-cron signature check by computing a valid hash using the publicly known constant salt embedded in the plugin source. Attackers can send forged GET requests with attacker-controlled parameters to the unauthenticated init-hooked endpoint to repeatedly trigger product-feed generation, causing CPU/database resource exhaustion and overwriting of the generated feed file.1dCVE-2026-133615——
——0A vulnerability has been identified in webauthn-kotlin versions 1.1.0 through 1.1.3, where an exported device-credential authentication activity launches attacker-supplied intents without validation, potentially allowing a malicious application on the same Android device to access application-protected data, depending on the host application’s content-provider configuration.11hCVE-2026-133613——
——0Flowintel's community data launcher script passes the admin API key as a command-line argument to a child Python helper process. On Unix-like systems, process arguments are readable by all local users via standard tools such as ps or by reading /proc/<pid>/cmdline, allowing any local user to observe the key while the helper process is running.
An attacker with local access to the host can capture the admin API key during the brief window in which the helper process is active. Possession of the key grants full administrative access to the Flowintel REST API, enabling unauthorized creation, modification, or deletion of threat intelligence data, users, roles, and organisations.
Preconditions:
- The victim must invoke the community data import or deletion command (launch.sh -tdc or -dtdc) with the API key supplied as a positional argument.
- The attacker must have a local user account on the same host and be able to observe process listings during the helper execution window.
Impact:
- Disclosure of the Flowintel admin API key to any local user.
- Subsequent full administrative control over the Flowintel instance via its REST API.
Affected: <4.1.012hCVE-2026-242056.4 MÉD—
——0NVIDIA TensorRT-LLM for any platform contains a vulnerability where a user may cause a race condition by running concurrent requests. A successful exploit of this vulnerability may lead to data tampering.3hCVE-2026-650949.0 CRÍ—
——0NVIDIA VIRTIO-Net for Bluefield contains a vulnerability in Virtio-Net where a VM user may cause a Write-What-Where condition by crafted message. A successful exploit of this vulnerability may lead to Code Execution in Virtio-Net scope.2hCVE-2026-1133477.2 ALT—
——0Multiple products produced by Ruijie Networks Co., Ltd. contain an OS command injection vulnerability that may lead to arbitrary OS commands being executed by an attacker with the administrative privilege.3hCVE-2026-133611——
——0vulnerability-lookup prior to version 7.1.0 contains two security issues in its web interface. First, the /user/toggle_2fa endpoint accepted HTTP GET requests and disabled the authenticated user's two-factor authentication without requiring password confirmation. Because the state-changing operation was reachable via GET, an attacker with a basic account could embed an image element (or any other GET-triggering markup) in a comment pointing to that endpoint. When a victim with 2FA enabled viewed the comment, their browser silently issued the GET request, disabling the victim's second factor of authentication. No CSRF token was enforced on the GET path, and no password was required to confirm the action. Second, the comment detail page (/comment/<uuid>) rendered unpublished comments to any visitor who possessed the URL, bypassing the moderation workflow that was intended to keep such content hidden until approved. The API correctly restricted access, but the web view did not enforce the same authorization check. The combined effect allows an attacker to degrade a victim's account security (removal of 2FA) through a single page view and to read content that was not yet approved for public release.2hCVE-2026-1335614.3 MÉD—
——0Mattermost versions 11.2.x <= 11.2.4 fail to validate that the redirect target on the web app landing page that prompts users to open links in the desktop app stays on the Mattermost origin which allows an unauthenticated attacker to redirect a user to an arbitrary external site via a crafted link. Mattermost Advisory ID: MMSA-2026-0089312hCVE-2026-1335055.9 MÉD—
——0When message history is serialized and loaded again, a tool result that contains a JSON object in the serialized shape of one of Pydantic AI's file types is rebuilt as that file. A third party who controls data a tool returns, such as a record that an MCP server or a function tool returns from an external system, can use this to make the application download a URL on its local network, or make the model provider read an s3:// object or uploaded file with the application's credentials. Applications are affected when tool results can contain third-party data, message history containing those results is serialized and loaded again (stored history, history a UI adapter holds on the server, or Temporal, DBOS or Prefect durable execution), and the model is one Pydantic AI downloads files for, one that honors force_download, or one whose provider resolves non-HTTP file URLs or uploaded files. History that a client submits through the Vercel AI or AG-UI adapter in the default mode is not affected. The issue is fixed in pydantic-ai and pydantic-ai-slim 1.107.10 and 2.66.0.2hCVE-2026-1330986.8 MÉD—
——0In JetBrains TeamCity before 2026.2.2 stored XSS in GitHub issue tracker suggestions was possible4hCVE-2026-1330977.4 ALT—
——0In JetBrains TeamCity before 2026.2.2, 2026.1.6 agent Push authorization keys were predictable6hCVE-2026-7731210.0 CRÍ—
——0Flyto2 Core is an execution kernel for automation and AI-agent workflows. Prior to 2.26.8, the earlier CVE-2026-67429 remediation applied validate_path_with_env_config() to only a hand-picked set of file-writing modules, leaving data.csv.write in src/core/modules/atomic/data/csv_write.py, data.json_to_csv in src/core/modules/atomic/data/json_to_csv.py, file.copy in src/core/modules/atomic/file/copy.py, and file.move in src/core/modules/atomic/file/move.py outside the FLYTO_SANDBOX_DIR confinement boundary. The advisory also identifies unguarded modules across the broader data.*, file.*, and report-writer families. data.csv.write relies on a .. substring denylist that does not reject absolute paths, while the other named modules accept caller-controlled output or destination paths without the canonical guard. A caller able to execute modules or recipes can therefore write attacker-controlled content outside the configured sandbox, potentially overwriting application code, recipes, startup files, scheduled-task files, or authorization files and gaining code execution on the host. This issue is fixed in version 2.26.8.22hCVE-2026-773158.5 ALT—
——0Flyto2 Core is an execution kernel for automation and AI-agent workflows. Prior to 2.26.8, http.paginate in src/core/modules/atomic/http/paginate.py validates only the initial base_url, and _paginate_link_header then follows a server-supplied Link: rel="next" URL without calling the SSRF guard again. A low-privileged caller can make an allowed first page point the next request at an internal or metadata service, causing a readable second-order SSRF whose response body is aggregated into all_items and returned. This issue is fixed in version 2.26.8.2hCVE-2026-1226768.5 ALT—
——0Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection') vulnerability in PublishPress PublishPress Authors publishpress-authors allows Blind SQL Injection.This issue affects PublishPress Authors: from n/a through 4.18.0.2hCVE-2026-1330967.1 ALT—
——0In JetBrains TeamCity before 2026.2.2, 2026.1.6 improper authorization allowed queuing builds with personal changes of other users4hCVE-2026-1333276.3 MÉD—
——0A vulnerability was identified in Dolibarr ERP CRM up to 23.0.4. This affects the function dol_convert_file in the library htdocs/core/lib/files.lib.php of the component Document Upload. Such manipulation leads to code injection. The attack can be launched remotely. The exploit is publicly available and might be used. Upgrading to version 24.0.0 is able to mitigate this issue. The name of the patch is 9b6368b2a9eae3eec690b2ce22c4f291f0051540. Upgrading the affected component is advised.21hCVE-2026-1330918.0 ALT—
——0In JetBrains TeamCity before 2026.2.2, 2026.1.6 stored XSS on the Users page was possible4hCVE-2026-1330907.2 ALT—
——0In JetBrains TeamCity before 2026.2.2, 2026.1.6 improper authorization allowed overwriting versioned settings of other projects6hCVE-2025-714736.1 MÉD—
——0lwext4 through 1.0.0 contains a heap buffer overflow in ext4_fs_init_inode_bitmap() in src/ext4_fs.c because it does not validate s_inodes_per_group against the inode bitmap size. Attackers can supply a crafted ext4 image with an oversized s_inodes_per_group so memset writes past the heap buffer, corrupting memory and crashing the mounting application.19hCVE-2026-1334995.5 MÉD—
——0A vulnerability was found in Vahe1994 AQLM up to e79a896ed6656fe4ed06193d42d004e7d0bbdbb2. The affected element is an unknown function of the file inference_kernels/triton_kernel.py of the component Triton Kernel. Performing a manipulation of the argument codes results in out-of-bounds read. It is possible to initiate the attack remotely. The exploit has been made public and could be used. This product is using a rolling release to provide continious delivery. Therefore, no version details for affected nor updated releases are available. The project was informed of the problem early through an issue report but has not responded yet.3hCVE-2026-1334986.3 MÉD—
——0A flaw has been found in mozilla-ai llamafile up to .0.10.5. This issue affects the function llamafile_open_zip of the file llamafile/llamafile.c of the component Container Parser. This manipulation of the argument size causes out-of-bounds read. The attack is possible to be carried out remotely. Upgrading to version 0.10.6 is capable of addressing this issue. Patch name: 5ac1c4dbb76737fee71a54922df95342c667ba6f. It is advisable to upgrade the affected component.2hCVE-2026-1334974.3 MÉD—
——0A vulnerability was detected in itsourcecode Banking Management System 1.0. This vulnerability affects unknown code of the file navbar.php. The manipulation of the argument page results in cross site scripting. The attack can be executed remotely. The exploit is now public and may be used.6hCVE-2026-1331092.4 BAI—
——0In JetBrains TeamCity before 2026.2.2, 2026.1.6 hTML injection in pipeline notification emails was possible via user and job names22hCVE-2026-1331082.7 BAI—
——0In JetBrains TeamCity before 2026.2.2 the current request details debug endpoint exposed authentication headers and session cookies22hCVE-2025-714746.1 MÉD—
——0lwext4 through 1.0.0 contains an out-of-bounds write vulnerability in ext4_fs_init_block_bitmap() that allows attackers to corrupt heap memory via crafted superblock geometry fields. Attackers can supply an ext4 image with oversized s_reserved_gdt_blocks and BLOCK_UNINIT groups to write past the bitmap buffer, crashing applications that mount it.6hCVE-2026-126540——
——0Convert::ASN1 versions before 0.37 for Perl allow CPU exhaustion via quadratic bigint loop in os2ip.
Every INTEGER or ENUMERATED longer than four content octets reaches os2ip, which iterates over each content octet with $result = ($result * 256) + $octet. Each multiplication operates on a bigint that grows by one octet per step, so decoding an n-octet value takes O(n^2) bigint work. Nothing limits content length before this loop runs.
An attacker who controls the decoded data can exhaust CPU with a single oversized INTEGER or ENUMERATED value.19hCVE-2026-126568——
——0Convert::ASN1 versions before 0.37 for Perl allow memory exhaustion via unbounded MSG_PEEK buffer doubling in asn_recv.
asn_recv peeks the socket in a loop, doubling the buffer size on each pass until a tag and length can be decoded. An inverted boolean guard ($tb || rather than $tb &&) prevents the need-more-data exit from firing once a tag byte has been decoded. When asn_decode_length returns no value because the length is absent or incomplete, no exit fires for that case either. MSG_PEEK leaves socket data unconsumed, so each re-peek returns at once while the buffer size doubles.
A single truncated BER header sent to a process calling asn_recv doubles the peek buffer until memory is exhausted. Perl's realloc failure is fatal and not catchable with eval.19hCVE-2026-126576——
——0Convert::ASN1 versions before 0.37 for Perl allow CPU exhaustion via unchecked exponent length in binary REAL decoding.
Binary REAL decoding never bounds-checks the exponent length against the available content. For a minimal one-byte binary REAL, the exponent byte is read from outside the element, making the mantissa length -1. Perl reads a negative length as an offset from the end of the string, so the mantissa is nearly all of the input that follows the element. The long-form exponent path also omits the extra length octet when sizing the mantissa.
An attacker who controls the decoded data can exhaust CPU with a list of such values, as each mantissa loop iterates over the remaining input. The decoded values are also computed from bytes outside each element.19hCVE-2026-126682——
——0Convert::ASN1 versions before 0.37 for Perl allow CPU exhaustion via exponent-notation string in INTEGER encoding.
The encoder routes any value with abs >= 2**31 to i2osp without validating that it is an integer. A string in exponent notation such as "1e8000" passes that threshold. Math::BigInt expands it to an 8001-digit integer, and i2osp's per-octet conversion loop is quadratic in that size.
An attacker who controls encoded INTEGER values can exhaust CPU with a short exponent-notation string.19h