Just published.
The newest CVEs, straight from CVE.org and NVD.
-
CVE-2026-77283
obsidian-sync-mcp is an MCP server that gives AI agents access to Obsidian vaults. Prior to 0.5.7, deployments running without MCP_AUTH_TOKEN install no serverOptions.authenticate check for the HTTP /mcp endpoint, so requests are dispatched without validating the Host or Origin headers. A malicious website visited by the operator can use DNS rebinding to reach loopback, LAN, VPN, or container-published addresses, or directly issue a cross-origin request to loopback, and invoke the MCP tool surface to read, create, overwrite, delete, or move vault notes. Deployments using MCP_AUTH_TOKEN are not affected because browsers do not attach the bearer token. This issue is fixed in version 0.5.7.
-
CVE-2026-129316
Dokploy is a free, self-hostable Platform as a Service (PaaS). From 0.29.13 until 0.31.0, the Docker terminal, container-log, and monitoring WebSocket handlers accept a caller-controlled serviceId as authorization but do not consistently bind the requested containerId or appName to that service. A member with read access to any service can consequently target unrelated containers, obtain a root shell in the Dokploy panel container, read other tenants' logs and statistics, and use the mounted Docker socket to compromise the host. This issue is fixed in version 0.31.0.
-
CVE-2026-129315
Dokploy is a free, self-hostable Platform as a Service (PaaS). In 0.31.0 and earlier, the hosted installer and manual installation instructions set /etc/dokploy to mode 0777, allowing a local non-root user on a shared host to replace root-owned schedule or Traefik configuration directories. Dokploy subsequently consumes those paths as trusted configuration and can execute substituted scheduled-task scripts as root, resulting in local privilege escalation.
-
CVE-2026-129990
Dokploy is a free, self-hostable Platform as a Service (PaaS). Prior to 0.31.1, every network.* procedure is only authentication-protected and does not require the Docker permission, allowing a standard member to list, inspect, create, import, recreate, resynchronize, and delete Docker networks. A member can disrupt service connectivity or alter network configuration, and create/import operations can target an unchecked serverId. This issue is fixed in version 0.31.1.
-
CVE-2026-125353
An incorrect permission assignment vulnerability in Omada SDN Controller prior to 6.2.20 may cause a database backup generated during the uninstall workflow to remain readable by other local users on the host. Successful exploitation requires local, low-privileged access to the affected host and a retained uninstall backup. An attacker meeting these conditions could access the backup and disclose controller data without controller credentials.
-
CVE-2026-75518
DOMShell is a Chrome automation project whose MCP server exposes Unix-style tools for AI agents. Prior to 2.0.8, the DOMShell MCP server classifies the domshell_eval tool and eval command as read-tier operations that can run without --allow-write, even though eval shares the Runtime.evaluate dispatch path with the write-tier js command and has no side-effect gate. An MCP caller with tool-session access can therefore use domshell_eval to mutate DOM, window, or global state while the host approval interface, server authorization flags, denial messages, and audit records represent the operation as read-only. The bypass does not expose a capability beyond domshell_js for callers already authorized to write, but it violates the configured permission boundary for callers lacking that authorization. This issue is fixed in version 2.0.8.
-
CVE-2026-132644
MISP contains an authorization flaw in the GalaxyCluster model's captureCluster() function, which handles the import and synchronization of Galaxy Clusters. Prior to the fix, any authenticated user holding the galaxy editor role could replace an existing locked Galaxy Cluster by supplying its UUID in an import request, regardless of which organization owned the cluster. The function unconditionally called captureOrganisationAndSG() before the save operation, overwriting the cluster's org_id and orgc_id fields with the importing user's organization, effectively transferring ownership of the cluster to the attacker's organization. Additionally, the save operation was not restricted to a specific set of fields, allowing arbitrary field modification. The vulnerable behavior required the attacker to possess a valid MISP account with the perm_galaxy_editor role and to know or discover the UUID of a target cluster belonging to another organization. No site-admin or owner-organization privilege was enforced for the update path. The result was that a low-privilege galaxy editor in one organization could modify the content (value, description, distribution, etc.) of clusters owned by other organizations and seize ownership, corrupting the integrity of shared threat-intelligence data across the MISP instance. The vulnerability was reported by ENISA and CERT.pl. The fix introduces an explicit authorization gate that restricts cluster updates to the owner organization, a sync user who can see the cluster via sharing-group rules, or a site administrator, and enforces a field allow-list that excludes org_id and orgc_id from update writes, preserving original ownership.
-
CVE-2026-132645
MISP contains an authorization bypass vulnerability in the Galaxy import/sync functionality (captureGalaxy in app/Model/Galaxy.php). The captureGalaxy() method matched an incoming galaxy payload to an existing galaxy by UUID and unconditionally overwrote the matched record, regardless of whether the importing user belonged to the owning organization or whether the galaxy was a system-provided default galaxy. Any authenticated user holding the perm_galaxy_editor role could therefore: (1) demote a default galaxy by setting its default flag to false, (2) reassign the galaxy's org_id and orgc_id to the attacker's organization, effectively taking ownership, and (3) modify other galaxy fields (name, description, distribution, version). Additionally, the role-level permission check (perm_site_admin or perm_galaxy_editor) was executed after the captureOrganisationAndSG() call, meaning organization and sharing-group data was already captured before the authorization decision was made. The vulnerability requires an authenticated user with the perm_galaxy_editor role and the ability to submit a galaxy import payload containing a UUID that matches an existing galaxy. The security impact is loss of data integrity across the MISP instance: default galaxy definitions are corrupted, cross-organization ownership is silently transferred, and the integrity of shared threat-intelligence galaxy metadata is compromised for all users of the instance.
-
CVE-2026-77297
MyMangaDB is a self-hosted manga collection manager with automatic metadata, MyAnimeList import, Docker support, and collection statistics. Prior to 2.7.2, the unauthenticated /api/v1/database/import handler in backend/app/api/v1/endpoints/database.py accepts a ZIP archive and moves an included database file to DATABASE_PATH while also replacing IMAGE_PATH content. An attacker can supply a database containing an attacker-controlled users table and administrative account, then authenticate as that account; invalid replacement data can also destroy application data or availability. This issue is fixed in version 2.7.2.
-
CVE-2026-73969
AnsibleForms is a self-hosted web application that turns automation into self-service forms. Prior to 6.2.1, AnsibleForms uses a public hard-coded fallback JWT signing secret from server/config/auth.config.js when ACCESS_TOKEN_SECRET is unset. An unauthenticated attacker can sign a self-contained token with privileged claims because server/src/auth/auth_jwt.js verifies the signature without reloading the user from the database, and server/src/lib/middleware.js trusts the token's roles and options for authorization. A forged token can therefore be accepted as an administrator session, resulting in authentication bypass and full privileged API access. This issue is fixed in version 6.2.1.
-
CVE-2026-69138
djangocms-versioning is a general-purpose versioning package for django CMS 4 and later. Prior to 2.6.1, the check_discard and check_edit_redirect Conditions on djangocms_versioning.models.Version omit user_can_change, while VersionAdmin.discard_view and VersionAdmin.edit_redirect_view obtain a version by its primary key and rely on those conditions for object-level authorization. With CMS_PERMISSION set to True, the affected views are gated only by is_staff, allowing any staff user to select content without object-level change permission. With the default CMS_PERMISSION value of False, a staff user must also hold the global Django change_ permission, but that model-level permission does not enforce the intended per-object authorization, so the user can delete a draft version and its content object through discard or create an editable draft from a published version through edit_redirect even when the user cannot change that specific content object. This issue is fixed in version 2.6.1.
-
CVE-2026-69131
django-filer is a file and image management application for Django. Prior to 3.5.0, the delete_files_or_folders action in filer/admin/folderadmin.py checks global Django model-level deletion permission but omits check_folder_edit_permissions and check_files_edit_permissions when FILER_ENABLE_PERMISSIONS is enabled. An authenticated staff user who has a global delete_file, delete_folder, or delete_image permission but is limited to read-only access by FolderPermission can therefore delete files, folders, and descendants from the restricted folder. Deployments using the default disabled per-folder permission system are not affected, and the issue does not create a read-confidentiality impact. This issue is fixed in version 3.5.0.
-
CVE-2026-132637
MISP contains an authorization bypass in the correlations/eventCorrelations endpoint and the underlying correlation engines (Default and OnDemand). The CorrelationsController::eventCorrelations action did not verify that the requesting user was permitted to view the event specified by the event ID parameter. Any authenticated user could supply the ID of an event belonging to another organisation (distribution level 0, organisation-only) and receive its full correlation results, including attribute identifiers and values. Additionally, for events the user could partially view, the correlation engines filtered only the related (target) side of each correlation row and did not vet the queried event's own source attributes. As a result, correlations keyed by attributes hidden from the user (organisation-only attributes, attributes in sharing groups the user is not a member of, or attributes inside organisation-only objects) were returned with their IDs and values. The Default engine's related-event listing (used in events/view) also included events related solely through such hidden attributes. The vulnerability was reported by ENISA. Successful exploitation requires a valid authenticated session with any role and knowledge or enumeration of a target event ID. The security impact is disclosure of threat-intelligence attribute values (IP addresses, domain names, etc.) and attribute identifiers that the user is not authorised to see, as well as confirmation of the existence of otherwise invisible events.
-
CVE-2026-132144
MISP contains an authorization flaw in the user login-email change logic. When an external authentication backend is configured (LDAP, Shibboleth, ApacheSecureAuth, or CustomAuth), the login email address serves as the key by which the directory service resolves a user's identity, organization, and role on each login. The __canChangeLogin() method in UsersController only checked the MISP.disable_user_login_change setting and site-admin status; it did not account for whether an external authentication backend was active. As a result, any authenticated non-admin user (including org admins) could change their own login email to an address that matches a higher-privilege account in the external directory. On the user's next login, the external backend would resolve that email to the target directory identity, granting the attacker the organization and role associated with that identity. This constitutes an account-identity capture and privilege escalation: the attacker effectively assumes the identity of any directory account whose email they can guess or discover. The vulnerability requires an authenticated session with a valid API key or session token and a MISP instance configured with at least one external authentication backend. The impact is limited to the scope of the captured directory identity's privileges within MISP.
-
CVE-2026-132137
A vulnerability was identified in Relaticle up to 3.10.0. This affects the function mapToField of the file packages/ImportWizard/src/Livewire/Steps/MappingStep.php of the component Import Wizard. Such manipulation of the argument Target leads to authorization bypass. The attack can be executed remotely. The exploit is publicly available and might be used. The vendor was contacted early about this disclosure.
-
CVE-2026-130090
When two templated upstreams in temporal-proxy render the same address, a request routed to one upstream can be sent under the other upstream's credential, namespace translation, and payload encryption keys. temporal-proxy pools connections for templated upstreams by rendered address and TLS server name only, although each upstream's outbound credential, namespace translation, and payload encryption keys, where configured, are attached to its connection. A request routed to one upstream can therefore be sent over the connection the other created, and the upstream Temporal service authorizes it as the other upstream's principal. Configuration validation rejects only identical hostPort values, so templates that differ in text but render the same address are accepted. This mixes the two upstreams' ordinary traffic, and a caller that an external authenticator restricts to one upstream's namespaces can use it to act under the other upstream's credential. Deployments with at most one templated upstream are not affected. temporal-proxy versions 0.1.0 through 1.1.1 are affected. The issue is fixed in 1.0.2, 1.1.2, and 1.2.0, and releases before 1.0.0 receive no fix. Upgrade to the fixed release for the deployed minor release line.
-
CVE-2026-127222
SGLang through 0.5.23, fixed in commit 7b5fb39, contains a missing authorization vulnerability in the POST /load_lora_adapter_from_tensors HTTP endpoint, which lacks the @auth_level(AuthLevel.ADMIN_OPTIONAL) decorator applied to its sibling admin endpoints, allowing unauthenticated network-reachable attackers to invoke it without an API key. Attackers can load a LoRA adapter from attacker-supplied serialized tensors to manipulate the served model, and when combined with a separately documented deserialization bypass, achieve unauthenticated remote code execution as root.
-
CVE-2026-111516
Dell Alienware Command Center (AWCC), versions prior to 6.14.105.0, contain a Least Privilege Violation vulnerability. A low privileged attacker with local access could potentially exploit this vulnerability, leading to Elevation of Privilege and Denial of Service.
-
CVE-2026-111514
Dell Alienware Command Center (AWCC), versions prior to 6.14.105.0, contain an Incorrect Default Permissions vulnerability. A low privileged attacker with local access could potentially exploit this vulnerability, leading to Arbitrary Code Execution and Elevation of Privilege.
-
CVE-2026-76807
Todoist MCP Server provides tools that allow AI agents to access and modify a Todoist account. Prior to 10.3.3, the optional todoist-mcp-http server accepts browser-origin requests to POST /mcp without validating the Host and Origin headers or authenticating the browser caller. A malicious website can use DNS rebinding to reach the loopback listener and invoke MCP tools through StreamableHTTPServerTransport, which uses the server-held Todoist API key. This exposes Todoist account data and permits operations such as task creation, updates, completion, and deletion. The stdio transport is not browser-reachable and is not affected by this attack path. This issue is fixed in version 10.3.3.
-
CVE-2026-132136
A weakness has been identified in Directus up to 12.6.1. Affected by this issue is the function getSecret of the file api/src/services/shares.ts of the component Share Login Service. This manipulation causes incorrect authorization. It is possible to initiate the attack remotely. The exploit has been made available to the public and could be used for attacks. The vendor was contacted early about this disclosure.
-
CVE-2026-45142
Chamilo LMS is an open-source learning management system. Prior to 1.11.40 and 2.0.1, main/lp/lp_ajax_save_objectives.php and public/main/lp/lp_ajax_save_objectives.php trust the request uid parameter instead of the authenticated user identity when saving SCORM learning objectives. Any authenticated student can replace uid with another user's identifier and overwrite that user's objectives because the endpoint performs no ownership check. This issue is fixed in versions 1.11.40 and 2.0.1.
-
CVE-2026-130785
Incus from 7.3.0 before 7.6.0 contains an authentication bypass vulnerability in the /dev/incus guest API that allows a process in one container to impersonate another by creating a cgroup named with a forged lxc.monitor. component. Attackers can read the target container's user.* and cloud-init.* configuration and devices, subscribe to its events, mark it ready, and export images across projects.
-
CVE-2026-130787
Frappe ERPNext before 16.35.0 contains a missing authorization vulnerability in financial reporting endpoints that allows low-privilege authenticated users to generate reports outside their permitted scope. Attackers can send crafted requests to affected reporting endpoints to read other companies' tax and payment data, including personal and corporate identifiers and financial totals.
-
CVE-2026-130789
SemanticMediaWiki from 7.3.0 before 7.4.0 contains a missing authorization vulnerability in the smwtask API check-query task that allows unauthenticated attackers to bypass object-level authorization by relying on the wiki-wide edit right. Attackers can supply arbitrary subject titles, including protected pages, to run synchronous Semantic MediaWiki queries with caller-controlled conditions, limit, offset, and querymode parameters.
-
CVE-2026-130782
Incus before 7.6.0 contains an improper authorization vulnerability that allows restricted and view-only clients to read default project resources by requesting the default project directly. Attackers knowing resource names can read profiles, images, network zone peer keys and stored credentials, and export custom volume or bucket backups to obtain full contents or S3 admin keys.
-
CVE-2026-130784
Incus before 7.6.0 contains an authorization bypass vulnerability that allows restricted clients to create and access default project resources by exploiting entitlement checks against the requested project. Attackers can create bridge networks with raw.dnsmasq directives to execute code as root on the host, or manipulate inherited resources via the ?project= parameter.
-
CVE-2026-130778
Incus before 7.6.0 contains an authentication bypass vulnerability that allows remote attackers to authenticate as any trusted certificate by appending its public certificate to their presented TLS client chain. Attackers can append the server's or a cluster member's certificate to gain cluster-level access, bypass authorization, modify trust store entries, and influence scriptlet-based authorization decisions.
-
CVE-2026-130774
Incus before 7.6.0 contains a privilege escalation vulnerability: rebuilding a never-started unprivileged container makes it boot without a user namespace. Root processes inside the container run as host uid 0, keeping capabilities like sys_module, with no device cgroup restriction, letting them load kernel modules and compromise the host.
-
CVE-2026-130775
Incus before 7.6.0 contains an authorization bypass vulnerability in the /dev/incus guest-API image export handler that fails to pin the project parameter to the guest's own project. Container root users with the guest-API image entitlement can supply another project in the project query parameter to export public or cached images, including non-public images pulled from authenticated remotes.
-
CVE-2026-130772
Incus before 7.6.0 contains an improper authorization vulnerability that allows restricted TLS clients to read other projects' bucket S3 keys and profile configuration through recursive and all-projects responses. Attackers can query bucket listings with recursion=2 or profiles with all-projects=true to obtain S3 secret keys and sensitive environment, cloud-init and user values across projects.
-
CVE-2026-130773
Incus before 7.6.0 contains an authorization bypass that allows restricted project users to create resources in the unrestricted default project by targeting projects lacking features.* flags. Attackers can create bridge networks with raw.dnsmasq dhcp-script injection to run scripts as root, or custom volumes with block.create_options mkfs arguments to read host files.
-
CVE-2026-130755
Froxlor before 2.3.16 contains an authorization bypass vulnerability that allows authenticated customers to read hidden traffic statistics by calling the Traffic.listing API command. Attackers can authenticate to api.php with their API key and secret to retrieve HTTP, FTP, and mail traffic volumes despite the panel.customer_hide_options restriction.
-
CVE-2026-130758
Fleet before 4.92.0 contains an authorization bypass vulnerability in the deprecated POST /api/latest/fleet/mdm/apple/enqueue endpoint that allows team-scoped users to enqueue Apple MDM commands outside their teams. Attackers with team admin, maintainer, or GitOps roles can include one own-team host UUID alongside unresolved enrollment IDs to send commands like InstallProfile to deleted or user-channel enrollments without activity logging.
-
CVE-2026-130751
OpenWrt LuCI luci-app-travelmate on master before 3.0.0-1 and on openwrt-25.12 before 2.4.9-6 grants file exec permissions for state-mutating commands in the read scope of its rpcd ACL, letting authenticated read-only users run those commands as root. Attackers can invoke ubus file exec on /etc/init.d/travelmate setup, start, stop, restart, or /sbin/ifup to modify and commit network and firewall configuration or disrupt the travelmate daemon.
-
CVE-2026-130752
OpenWrt LuCI luci-app-uhttpd before commit d38658bedbdd811b798da16b49301478c6cf21af grants '/*' file read in its rpcd read ACL, which rpcd's fnmatch matching applies to every path. Users delegated only luci-app-uhttpd read access can read any file as root, including /etc/shadow, /etc/config/* credentials, /etc/uhttpd.key, and /etc/dropbear keys.
-
CVE-2026-130754
All versions of LaraDashboard contain an authentication bypass in SocialAuthService::findOrCreateUser(), which links a social login to an existing account by email without checking that the OAuth provider verified that email. Unauthenticated attackers using an OAuth/OIDC provider that asserts a victim's email can log in as any user, including Superadmin, and keep persistent access through a linked social_accounts entry.
-
CVE-2026-128664
An insecure password reset vulnerability in the Admin Password Recovery Module of PHPGurukul News Portal Project v4.1 allows unauthenticated remote attackers to change an administrator or sub-administrator account password by providing the corresponding username and email address. The application does not require verification of account ownership through an email token, OTP, or another authentication mechanism before updating the password, potentially resulting in account takeover.
-
CVE-2026-132229
A flaw was found in hypershift-addon. Due to an overly permissive role-based access control (RBAC) configuration that grants permissions to all authenticated users rather than restricting them to the addon service, a low-privileged authenticated user can view, modify, or delete managed cluster resources. This improper authorization allows unauthorized users to alter or disrupt clusters across the entire managed fleet.
-
CVE-2026-86403
Mattermost versions 11.11.x <= 11.11.0, 11.7.x <= 11.7.11, 11.10.x <= 11.10.2 fail to check whether the calling session originated from an OAuth authorization grant, which allows a third-party application holding a delegated access token to register a new, independent OAuth application under the identity of the delegating user via the OAuth application registration endpoint.. Mattermost Advisory ID: MMSA-2026-00780
-
CVE-2026-89218
Mattermost versions 11.11.x <= 11.11.0, 11.7.x <= 11.7.11, 11.10.x <= 11.10.2 fail to enforce the Access Control Policy's download_file_attachment check across several post- and file-serving code paths (bulk post retrieval, channel bookmarks, and persistent notification delivery), which allows an authenticated channel member denied a specific file's download to obtain that file's metadata and a rendered preview thumbnail via those endpoints and events.. Mattermost Advisory ID: MMSA-2026-00785
-
CVE-2026-18079
Mattermost versions 11.11.x <= 11.11.0, 11.7.x <= 11.7.11, 11.10.x <= 11.10.2 Fail to revoke a user's active sessions when a bulk import links an existing user account to a bot record or changes the account's auth_service/auth_data, which allows an attacker in possession of a session token issued before the import to retain full authenticated access to the affected account via a system administrator-triggered bulk import job.. Mattermost Advisory ID: MMSA-2026-00746
-
CVE-2026-18396
Mattermost versions 11.7.x <= 11.7.11, 11.10.x <= 11.10.2, 11.11.x <= 11.11.1 fail to enforce playbook membership management permissions when creating a new playbook which allows a low-privileged user to grant the playbook_admin role to an arbitrary user via a crafted playbook creation request. Mattermost Advisory ID: MMSA-2026-00749
-
CVE-2026-18399
Mattermost versions 11.7.x <= 11.7.11, 11.10.x <= 11.10.2, 11.11.x <= 11.11.1 fail to enforce the configured GitLab group/namespace restriction on the permalink code-preview feature and the {{get_projects}} MCP tool, which allows an authenticated user with a connected GitLab account to retrieve repository file content or project metadata from projects outside the restricted group via a crafted GitLab permalink or MCP tool call. Mattermost Advisory ID: MMSA-2026-00750
-
CVE-2026-18476
Mattermost versions 11.7.x <= 11.7.11 fail to scope an incoming webhook's direct message target to the webhook's own team, which allows a team administrator to deliver a direct message to a user on a team they have no authority over via a webhook whose posting identity belongs to both teams.. Mattermost Advisory ID: MMSA-2026-00751
-
CVE-2026-18643
Mattermost versions 11.7.x <= 11.7.11 fail to enforce the team view permission across all filter combinations of the user listing endpoint which allows an authenticated user to enumerate the membership of a team they are not permitted to view via a crafted request to that endpoint.. Mattermost Advisory ID: MMSA-2026-00755
-
CVE-2026-17537
Mattermost versions 11.11.x <= 11.11.0, 11.7.x <= 11.7.11, 11.10.x <= 11.10.2 fail to enforce the channel Access Control Policy's file download permission check when serving file metadata and thumbnails through the drafts, scheduled posts, and post edit history endpoints, which allows an authenticated user who has lost eligibility under the policy to view metadata and thumbnails of file attachments via those endpoints, even though direct file access is correctly denied. Mattermost Advisory ID: MMSA-2026-00740
-
CVE-2026-18075
Mattermost versions 11.11.x <= 11.11.0, 11.7.x <= 11.7.11, 11.10.x <= 11.10.2 fail to invalidate cached WebSocket session state when an OAuth app is deleted, which allows a revoked OAuth integration to keep receiving real-time events, including private channel posts, via a WebSocket connection that was already open when the app was deleted.. Mattermost Advisory ID: MMSA-2026-00745
-
CVE-2026-132250
In JetBrains TeamCity before 2026.1.3 privilege escalation was possible via Basic authentication with a restricted access token
-
CVE-2026-132251
In JetBrains TeamCity before 2026.1.3 privilege escalation was possible by issuing new tokens from a restricted access token
The newest published CVE records from CVE.org and NVD, refreshed every few minutes, with severity, exploit odds and fix status.