Vulnerabilities exploitable today
376,275in current view
Single score combining CVSS, KEV membership and EPSS. Every CVE with its own record — timeline from publication to active exploitation.
In KEV catalog1,710
New KEV · 24H0
Exploit Today ≥ 701,646
Distribution · last window
- Critical2,387
- High8,799
- Medium6,801
- Low737
Filters
Window
Severity
Flags
CVECVSSEPSSKEVRExploitTitleMod.
CVE-2026-923634.3 MED—
———A flaw has been found in ag-ui-protocol ag-ui 1.0. Affected is an unknown function of the file src/stream/sse_parser.cpp of the component JSON Parser. Executing a manipulation can lead to resource consumption. The attack may be performed from remote. This patch is called ab6e0bc298996caac2b4b0b3ec0bd8d32a15a186. Applying a patch is advised to resolve this issue.18hCVE-2026-812408.6 HIG—
———Dell Wyse Management Suite, versions prior to 2605.0.3.683, contain an Unrestricted Upload of File with Dangerous Type vulnerability. An unauthenticated attacker with remote access could potentially exploit this vulnerability, leading to Remote execution.15hCVE-2026-812398.6 HIG—
———Dell Wyse Management Suite, versions prior to 2605.0.3.683, contain an Unrestricted Upload of File with Dangerous Type vulnerability. An unauthenticated attacker with remote access could potentially exploit this vulnerability, leading to Remote execution.15hCVE-2026-923646.3 MED—
———A vulnerability has been found in itsourcecode Leave Management System 1.0. Affected by this vulnerability is an unknown functionality of the file /module/employee/index.php. The manipulation of the argument ID leads to sql injection. It is possible to initiate the attack remotely. The exploit has been disclosed to the public and may be used.18hCVE-2026-923654.3 MED—
———A vulnerability was found in vllm-project vllm up to 0.29.0. Affected by this issue is some unknown functionality of the file vllm/v1/sample/thinking_budget_state.py. The manipulation results in inefficient algorithmic complexity. It is possible to launch the attack remotely. The pull request to fix this issue awaits acceptance.18hCVE-2026-616679.9 CRI—
———DIRAC is an interware, meaning a software framework for distributed computing. Prior to versions 8.0.79, 9.0.22, and 9.1.10, DataManagementSystem/Service/FileCatalogHandler.py checkDataset forwards an authenticated caller-controlled datasets value to DatasetManager.py __checkDataset, where datasetName is interpolated into an FC_MetaDatasets SQL query without parameterization. The injected query can control the returned MetaQuery value, which is passed to Python eval and permits command execution as the account running the DIRAC services. Successful exploitation can expose dirac.cfg, database passwords, stored proxies, and tokens, fully compromise the DIRAC system, and allow alteration of local log evidence. This issue is fixed in versions 8.0.79, 9.0.22, and 9.1.10.2dCVE-2026-196685.3 MED—
———A BIND recursive resolver may experience excessive resource consumption if it encounters large numbers of a particular kind of invalid DNSSEC record. Default limits on "max-records-per-type" and "max-types-per-name" help mitigate the exposure.
This issue affects BIND 9 versions 9.11.0 through 9.18.50, 9.20.0 through 9.20.27, 9.21.0 through 9.21.25, 9.11.3-S1 through 9.18.50-S1, and 9.20.9-S1 through 9.20.27-S1.17hCVE-2026-584857.1 HIG—
———mcp-searxng is a Model Context Protocol server that gives AI assistants web search and URL-reading capabilities through SearXNG. Prior to 1.7.1, web_url_read receives its caller-controlled URL through src/index.ts and validates only the literal hostname in assertUrlAllowed() within src/url-reader.ts before undiciFetch() performs operating-system DNS resolution. A public-looking attacker-controlled hostname that resolves to a private, loopback, link-local, or cloud-metadata address therefore passes the lexical check and causes the MCP server to connect to the internal destination. In the default HTTP configuration, an unauthenticated network client can use this path to read internal services, expose credentials or service tokens, and enumerate reachable internal hosts; in STDIO deployments, prompt-influenced tool selection can provide the malicious URL. Direct private IP literals are blocked, and MCP_HTTP_ALLOW_PRIVATE_URLS remains an explicit opt-out. This issue is fixed in version 1.7.1.2dCVE-2025-365914.4 MED—
———Dell ECS versions 3.8.1.0 through 3.8.1.7, and Dell ObjectScale versions prior to 4.4.0.0, contains an Use of a Broken or Risky Cryptographic Algorithm vulnerability. A high privileged attacker with local access could potentially exploit this vulnerability, leading to Information exposure.15hCVE-2026-924698.1 HIG—
———zlt2000 microservices-platform through 6.0.0 contains an authorization bypass vulnerability in the file-center module DELETE /files/{id} endpoint that performs no ownership validation. Authenticated attackers can enumerate file identifiers via GET /files and delete arbitrary users' files and metadata by supplying their identifiers to the delete endpoint.18hCVE-2026-182127.5 HIG—
———A flaw was found in the SAML Redirect Binding implementation of Keycloak, an open-source identity and access management solution. The issue occurs because the custom DEFLATE compression and decompression helpers fail to release native zlib memory after use. An unauthenticated attacker can exploit this by sending repeated malformed SAML requests, leading to native memory exhaustion and a denial of service.16hCVE-2026-196667.5 HIG—
———On a resolver configured to use ``dns64``, if an applicable answer from the authoritative server is malformed in a specific way, the resolver `named` process will exit unexpectedly.
This issue affects BIND 9 versions 9.11.0 through 9.18.50, 9.20.0 through 9.20.27, 9.21.0 through 9.21.25, 9.11.3-S1 through 9.18.50-S1, and 9.20.9-S1 through 9.20.27-S1.17hCVE-2026-57442——
———MCPVault is a lightweight Model Context Protocol server for safe access to files in an Obsidian vault. Prior to 0.11.5, PathFilter in src/pathfilter.ts uses root-anchored deny-list patterns, so nested .git, .obsidian, and node_modules path segments do not match the restriction and pass both isAllowed() and isAllowedForListing(). An attacker who influences a path selected by an AI agent can traverse nested repository or Obsidian metadata, read remote URLs or embedded tokens, or cause nested node_modules content to pollute the listAllTags index. This issue is fixed in version 0.11.5.21hCVE-2026-631278.2 HIG—
———RMCP is an official Rust SDK for the Model Context Protocol. Prior to 2.0.0, the rmcp crate's OAuth implementation in crates/rmcp/src/transport/auth.rs omits the RFC 9728 resource field from ResourceServerMetadata and allows discover_oauth_server_via_resource_metadata to use protected-resource metadata without confirming that the returned resource identifier exactly matches the configured MCP server. A malicious MCP server can publish metadata for a different legitimate MCP resource and its authorization server, causing a victim who connects and completes the authorization flow to obtain a legitimate access token that the client subsequently sends to the malicious server. The attacker can capture the token and impersonate the victim against the legitimate MCP resource within the token's granted scopes. This issue is fixed in version 2.0.0.20hCVE-2026-636718.1 HIG—
———MDC is a tool to take regular Markdown and write documents interacting deeply with a Vue component. Prior to 0.22.1, @nuxtjs/mdc uses parseMarkdown with allowDangerousHtml enabled by default and relies on validateProps, validateProp, and unsafeLinkPrefix to remove executable URLs from untrusted Markdown. validateProp checks only attributes named href or src, allowing an SVG xlink:href value represented as xLinkHref to retain a javascript: URL that executes in the page origin when selected. The data:text/html denylist entries are also compared against url.protocol, which is only data:, so an iframe src containing data:text/html survives sanitization and executes in an opaque origin when loaded. Plain href javascript: URLs, srcdoc, object, script, and base elements are already blocked, making these two paths specific sibling gaps in the sanitizer. This issue is fixed in version 0.22.1.21hCVE-2026-749098.1 HIG—
———Keycloak provides a policy enforcer to protect applications by matching incoming web requests against defined security policies. A flaw was found where the enforcer fails to correctly normalize web addresses that contain special encoded characters, such as those representing semicolons or directory traversal segments. An authenticated user can use these encoded characters to trick the enforcer into applying a less restrictive security policy than intended, potentially gaining unauthorized access to sensitive administrative or private application endpoints.16hCVE-2026-761637.5 HIG—
———If BIND is loaded with a "`named.conf`" file that contains no global "`options`" block, an attacker can send a query of QTYPE TKEY which may cause an assertion failure and subsequent unexpected program exit.
This issue affects BIND 9 versions 9.20.0 through 9.20.27, 9.21.0 through 9.21.25, and 9.20.9-S1 through 9.20.27-S1.17hCVE-2026-771195.9 MED—
———A validly signed NSEC3 from an unrelated sibling zone may be accepted as an insecurity proof, downgrading a secure delegation and letting a forged unsigned answer through.
This issue affects BIND 9 versions 9.11.0 through 9.18.50, 9.20.0 through 9.20.27, 9.21.0 through 9.21.25, 9.11.3-S1 through 9.18.50-S1, and 9.20.9-S1 through 9.20.27-S1.17hCVE-2026-774016.8 MED—
———Zope AccessControl provides a general security framework for use in Zope. Prior to 7.4, applications that allow untrusted users to create and execute AccessControl-controlled Python code do not safely guard str.format and str.format_map when those methods are reached through a str subclass. In both ImplPython.py and cAccessControl.c, Python formatting can recursively access attributes and subscriptions using unrestricted getattr and getitem behavior instead of the policy-restricted getattr and getitem operations. A controlled format string can therefore disclose objects reachable from values available to the formatting operation. This issue is fixed in version 7.4.17hCVE-2026-77409——
———RabbitMQ amqp091-go is a Go AMQP 0.9.1 client. Prior to 1.13.0, Channel.dispatch in channel.go, confirms.confirm in confirms.go, and Connection.dispatch0 in connection.go synchronously send publisher confirmations, flow-control events, consumer cancellations, returned messages, including NotifyConfirm events and connection block notifications, to application-provided channels. If a listener channel is unbuffered, full, or not drained promptly, the sole reader goroutine blocks and stops processing frames, acknowledgments, deliveries, and heartbeats. Broker-driven event bursts can therefore cause connection stalls, missed heartbeats, deadlocks, and disconnection. This issue is fixed in version 1.13.0.21hCVE-2026-77410——
———RabbitMQ amqp091-go is a Go AMQP 0.9.1 client. Prior to 1.13.0, Channel.recvContent in channel.go preallocates the message body slice with the uint64 ch.header.Size value supplied by an AMQP content header without capping the allocation to the negotiated Connection.Config.FrameSize value. A malicious or compromised broker can send an extreme declared body size and cause the Go runtime to attempt a correspondingly large allocation before body data is received. The allocation can exhaust memory and terminate the client process. This issue is fixed in version 1.13.0.20hCVE-2026-77411——
———RabbitMQ amqp091-go is a Go AMQP 0.9.1 client. Prior to 1.13.0, readLongstr in read.go returns an empty string and a nil error when a declared AMQP longstr length exceeds 0x7FFFFFFF instead of returning ErrSyntax. The function leaves the declared field bytes unread, while readTable treats the operation as successful and continues parsing from the wrong offset. A malicious or compromised broker can provide an oversized longstr in a table field and desynchronize subsequent AMQP parsing, causing attacker-controlled trailing bytes to be interpreted as later fields or frames and disrupting connection integrity and availability. This issue is fixed in version 1.13.0.17hCVE-2026-77412——
———RabbitMQ amqp091-go is a Go AMQP 0.9.1 client. Prior to 1.13.0, readField in read.go reads the length of an AMQP byte-array field with type tag x into a signed int32 and passes the value directly to make when allocating the field buffer. A malicious or compromised broker can encode a value such as 0xFFFFFFFF, which becomes -1 and causes a len out of range runtime panic. The panic escapes the network reader goroutine and terminates the client process, including during connection.start server properties or message header table parsing. This issue is fixed in version 1.13.0.20hCVE-2026-77407——
———RabbitMQ amqp091-go is a Go AMQP 0.9.1 client. Prior to 1.13.0, PlainAuth values defined in auth.go retain passwords as exported plaintext fields in Connection.Config.SASL after a successful PLAIN authentication handshake. The Connection.openComplete method in connection.go does not clear those values. Code with access to the Connection object, including reflective loggers, application performance monitoring agents, debugging utilities, and panic handlers, can traverse the configuration and expose the credentials to logs or state captures. The credential remains available for the lifetime of the connection instead of being cleared after authentication. This issue is fixed in version 1.13.0.21hCVE-2026-77408——
———RabbitMQ amqp091-go is a Go AMQP 0.9.1 client. Prior to 1.13.0, the writeShortstr function in write.go casts the byte length of AMQP shortstr property values to uint8 without first rejecting values longer than 255 bytes. An application that accepts an oversized CorrelationId, ReplyTo, MessageId, Expiration, UserId, AppId, ContentType, ContentEncoding, or Type value can therefore serialize a wrapped length and only a truncated prefix, while reporting no error. The resulting silent metadata corruption can break request and reply correlation, routing, tracing, and downstream message processing. This issue is fixed in version 1.13.0.21hCVE-2026-584837.5 HIG—
———mcp-searxng is a Model Context Protocol server that gives AI assistants web search and URL-reading capabilities through SearXNG. Prior to 1.7.1, web_url_read in src/index.ts passes a caller-supplied URL to readUrlContent() in src/url-reader.ts, where checkContentLength() treats a missing Content-Length header as an inconclusive preflight and the normal and error paths then consume the complete body with response.text(). A server that omits Content-Length can therefore bypass URL_READ_MAX_CONTENT_LENGTH_BYTES and force unbounded memory use. The resulting string is also processed by NodeHtmlMarkdown.translate(), increasing CPU consumption and allowing an unauthenticated HTTP client to cause denial of service. This issue is fixed in version 1.7.1.2dCVE-2026-829648.8 HIG—
———Improper preservation of permissions in the Avast sandbox minifilter driver (aswSnx.sys) on Windows allows a local, low-privileged attacker executing inside the sandbox to escape file isolation and escalate to SYSTEM.
When the sandbox virtualizes a file it copies the original security descriptor, but the driver opened the virtualization target object with GENERIC_WRITE and FILE_WRITE_ATTRIBUTES only, omitting WRITE_DAC. Every attempt to apply the original DACL therefore failed, and the failure was discarded silently, leaving virtualized copies of sensitive files with permissive permissions. Because the IRP_MJ_CREATE callback additionally did not strip WRITE_DAC for sensitive directories, a sandboxed process could rewrite the security descriptor of a virtualized object, read the virtualized copy of the SAM database, extract local NTLM password hashes and execute code as SYSTEM.
The absence of an IRP_MJ_SET_SECURITY callback in the driver's operation registration table is a related defense-in-depth gap, but it is not the control that prevents this attack.16hCVE-2026-848588.8 HIG—
———ScadaLTS 2.8.1-release-candidate build 0 is affected by an Authenticated Remote Code Execution via Scripting Sandbox Bypass
The DWR "DataSourceEditDwr" class exposes the "validateScript" method that compiles and executes attacker-supplied JavaScript via the Rhino scripting engine. There are no authorization checks on this method and so it is possible for an attacker with access to a low privilege user to abuse this flaw by leveraging the DWR routing bypass.21hCVE-2026-848596.5 MED—
———ScadaLTS 2.8.1-release-candidate build 0 is affected by an Authenticated Blind SQL Injection
The /api/events/search endpoint accepts a JSON body containing a sortBy array. The values in this array are concatenated directly into the SQL ORDER BY clause without any sanitization or parameterization. This allows authenticated users with the ROLE_USER role to perform time-based and boolean-based blind SQL injection to extract arbitrary data from the database, including password hashes of all users.
The endpoint is accessible to any authenticated user with ROLE_USER, ROLE_ADMIN, or ROLE_PUBLIC via POST /api/events/search as defined in spring-security.xml.21hCVE-2026-848608.8 HIG—
———ScadaLTS 2.8.1-release-candidate build 0 is affected by an Authorization Bypass
Spring Security gates DWR endpoints by URL path pattern, but DWR itself dispatches method calls based on the POST body parameters c0-scriptName and c0-methodName. The crossDomainSessionSecurity setting in web.xml is set to false, which disables DWR's built-in origin validation. This means any authenticated user can invoke any DWR method (regardless of the URL-based access control) by sending their request to a URL they are permitted to access (e.g. MiscDwr.initializeLongPoll.dwr) while targeting a restricted class in the POST body.
This is the systemic root cause that enables multiple other findings to be exploited as a low privilege user.21hCVE-2026-802747.5 HIG—
———If a BIND resolver sends a query for a DNSSEC-signed authoritative zone, and the authoritative server replies with a valid wildcard answer and signed NSEC3 proof, followed by an unsigned NSEC at the same owner name, it will trigger an unexpected program exit.
This issue affects BIND 9 versions 9.11.0 through 9.18.50, 9.20.0 through 9.20.27, 9.21.0 through 9.21.25, 9.11.3-S1 through 9.18.50-S1, and 9.20.9-S1 through 9.20.27-S1.17hCVE-2026-57441——
———MCPVault is a lightweight Model Context Protocol server for safe access to files in an Obsidian vault. Prior to 0.11.4, PathFilter in src/pathfilter.ts compiles restricted-directory patterns case-sensitively and compares paths without canonicalizing filesystem-equivalent segment names. On case-insensitive macOS and Windows filesystems, case variants of .git, .obsidian, or node_modules pass both isAllowed() and isAllowedForListing() even though the operating system opens the restricted directory, and Windows trailing dots or spaces provide the same bypass. An attacker who influences a path selected by an AI agent can use the bypass in read, write, move, search, or listing operations to expose or modify sensitive repository and Obsidian metadata. Vault-root .. containment is not affected. This issue is fixed in version 0.11.4.2dCVE-2026-880648.8 HIG—
———Backstage is an open framework for building developer portals. Prior to 1.14.6 and from 1.15.0 until 1.15.4, the @backstage/plugin-techdocs-node package insufficiently validates mkdocs.yml supplied by an authenticated user who can register or modify a TechDocs source. Unsafe Python YAML tags, markdown_extensions names and configuration, theme options, and extra_templates values can reach the documentation generator and cause unintended code execution. The resulting impact is limited to the files, credentials, network access, and other resources available to the TechDocs backend or build container. This issue is fixed in versions 1.14.6 and 1.15.4.20hCVE-2026-890315.4 MED—
———Adenion Blog2Social plugin for WordPress before 9.1.0 allows low-privileged users to modify the scheduled post records of other users. The b2s_calendar_move_post AJAX handler in includes/Ajax/Post.php issues an UPDATE against the b2s_posts table using only the attacker-supplied b2s_id primary key with no blog_user_id ownership constraint, allowing any user with the edit_posts capability to reschedule, suppress, or alter the publication state of any other user's scheduled social media post.21hCVE-2026-923834.3 MED—
———A security vulnerability has been detected in PbootCMS up to 3.2.24. This vulnerability affects the function UserController::del/UserController::mod of the file apps/admin/controller/system/UserController.php of the component User Management. Such manipulation leads to cross-site request forgery. The attack may be performed from remote. The exploit has been disclosed publicly and may be used. Upgrading to version 3.2.25 is able to resolve this issue. The name of the patch is c25241a0964742cefb7f698efbb6c38b868d6ff7. It is advisable to upgrade the affected component.18hCVE-2026-923959.1 CRI—
———@fastify/proxy-addr is a Fastify plugin that determines a request's client address behind trusted reverse proxies, and it backs Fastify request.ip and request.ips. In versions 3.0.0 through 5.1.0, a trust subnet written in IPv4-mapped IPv6 notation with an IPv4-sized prefix, such as ::ffff:10.0.0.0/8 instead of the correct ::ffff:10.0.0.0/104, is accepted without error but trusts every IPv4 address on the internet rather than the block it names. Because the socket peer then becomes trusted at hop 0, any unauthenticated client can supply an arbitrary X-Forwarded-For header and control the address the application reads, which defeats IP-based access control, rate limiting, geolocation, and audit logging. The plugin inherited this defect from the upstream proxy-addr module (CVE-2026-90711). The issue is fixed in @fastify/proxy-addr 5.1.1, and users should upgrade to 5.1.1 or later. As a workaround, ensure any IPv4-mapped IPv6 trust subnet uses a prefix length of at least 97, or express the range in plain IPv4 notation.16hCVE-2026-923807.3 HIG—
———A flaw has been found in WuzhiCMS up to 4.1.0. The impacted element is the function ckditor::saveRemote of the file coreframe/app/attachment/index.php of the component Remote Image Fetch. This manipulation of the argument source[] causes server-side request forgery. The attack can be initiated remotely. The exploit has been published and may be used. The project was informed of the problem early through an issue report but has not responded yet.18hCVE-2026-923813.5 LOW—
———A weakness has been identified in PbootCMS up to 3.2.22. This affects the function decode_string of the file apps/admin/controller/content/ContentController.php of the component Template Rendering. This manipulation of the argument Title causes cross site scripting. The attack is possible to be carried out remotely. The exploit has been made available to the public and could be used for attacks. The reported GitHub issue was closed with the reason "completed".18hCVE-2026-923667.3 HIG—
———A vulnerability was determined in code-projects Matrimonial System 1.0. This affects an unknown part of the file /search.php of the component Regular Search. This manipulation of the argument sex/mothertongue/maritialstatus/country/state/religion/agemin/agemax causes sql injection. The attack can be initiated remotely. The exploit has been publicly disclosed and may be utilized.16hCVE-2026-568306.5 MED—
———Shopper is a Headless e-commerce Admin Panel. Prior to 2.9.2, an earlier product sub-form hardening change left store() in packages/admin/src/Livewire/Components/Products/Form/Media.php without the edit_products authorization check used by sibling sub-forms. An authenticated staff user with browse_products can invoke the Livewire store action and replace the thumbnail and gallery images for a product whose Media component was initialized, even without product-edit permission. The product binding is locked, so the attacker cannot redirect the update to an arbitrary product through client-side ID substitution, and the impact is limited to products whose edit pages were loaded. This issue is fixed in version 2.9.2.2d