In open-webui version 0.3.8, the endpoint `/models/upload` is vulnerable to arbitrary file write due to improper handling of user-supplied filenames. The vulnerability arises from the usage of `file_path = f"{UPLOAD_DIR}/{file.filename}"` without proper input validation or sanitization. An attacker can exploit this by manipulating the `file.filename` parameter to include directory traversal sequences, causing the resulting `file_path` to escape the intended `UPLOAD_DIR` and potentially overwrite arbitrary files on the system. This can lead to unauthorized modifications of system binaries, configuration files, or sensitive data, potentially enabling remote command execution.

In version 0.3.8 of open-webui/open-webui, an arbitrary file write vulnerability exists in the download_model endpoint. When deployed on Windows, the application improperly handles file paths, allowing an attacker to manipulate the file path to write files to arbitrary locations on the server's filesystem. This can result in overwriting critical system or application files, causing denial of service, or potentially achieving remote code execution (RCE). RCE can allow an attacker to execute malicious code with the privileges of the user running the application, leading to a full system compromise.

In version v0.3.8 of open-webui/open-webui, the endpoint /api/pipelines/upload is vulnerable to arbitrary file write and delete due to unsanitized file.filename concatenation with CACHE_DIR. This vulnerability allows attackers to overwrite and delete system files, potentially leading to remote code execution.

An improper access control vulnerability in open-webui/open-webui v0.3.8 allows attackers to view and delete any files. The application does not verify whether the attacker is an administrator, allowing the attacker to directly call the GET /api/v1/files/ interface to retrieve information on all files uploaded by users, which includes the ID values. The attacker can then use the GET /api/v1/files/{file_id} interface to obtain information on any file and the DELETE /api/v1/files/{file_id} interface to delete any file.

An improper access control vulnerability in open-webui/open-webui v0.3.8 allows an attacker to view admin details. The application does not verify whether the attacker is an administrator, allowing the attacker to directly call the /api/v1/auths/admin/details interface to retrieve the first admin (owner) details.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the /api/generate, /api/embed, /api/embeddings, and /api/show endpoints accept any model name from the user and forward the request to the Ollama backend without checking whether the user is authorized to access that model. These endpoints only require get_verified_user (any authenticated non-pending user) and validate that the model exists in the full unfiltered model list, but never check AccessGrants.has_access(). This vulnerability is fixed in 0.9.0.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the channel router does not call filter_allowed_access_grants on either create or update paths. A non-admin user who can create group channels (or who owns a channel) can submit arbitrary access grants — including public wildcard grants — and those grants are stored verbatim, bypassing the admin's permission framework. This vulnerability is fixed in 0.9.0.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, any authenticated user with low privileges can enumerate active background tasks across the system and stop tasks belonging to other users via the GET /api/tasks and POST /api/tasks/stop/{task_id} methods. This allows a casual user to disrupt system-wide chat usage by continuously canceling other users' active tasks. This is a real authorization vulnerability affecting integrity and usability in multi-user deployments. This vulnerability is fixed in 0.9.0.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, Open WebUI supports model composition via base_model_id: a user-defined model (e.g., "Cheap Assistant") can reference an existing base model (e.g., "gpt-4-turbo-restricted") that provides the actual inference capability. When a user queries the composed model, the access control pipeline verifies the user has access to the composed model but never re-verifies access to the chained base model. Additionally, the model creation and import endpoints accept arbitrary base_model_id values without checking that the caller has access to that base model. Combined, this allows any user with the default model creation permission to create a model that chains to a restricted base model — and then invoke it, causing the server to dispatch the request to the restricted base model using the admin-configured API key. This vulnerability is fixed in 0.9.0.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the GET /api/v1/channels/{id}/members endpoint only checks membership for group and dm channel types (lines 467-469). For standard channels — including private ones — there is no channel_has_access check before returning the member list. Any authenticated user who knows a private channel's UUID can enumerate all users with access to that channel. This vulnerability is fixed in 0.9.0.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, he LDAP and OAuth authentication flows use a TOCTOU (Time-of-Check-Time-of-Use) pattern for first-user admin role assignment. The regular signup handler (signup_handler in auths.py, line 663) was explicitly patched to prevent this race with the comment "Insert with default role first to avoid TOCTOU race", but the LDAP and OAuth code paths were never updated with the same fix. This vulnerability is fixed in 0.9.0.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the POST /api/v1/retrieval/process/web endpoint accepts a user-supplied collection_name and an overwrite query parameter (default: True). It performs no authorization check on whether the calling user owns or has write access to the target collection. When overwrite=True, save_docs_to_vector_db calls VECTOR_DB_CLIENT.delete_collection() on the target collection before writing new content. This vulnerability is fixed in 0.9.0.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the type: "file" (non-full-context), type: "text" with collection_name, and bare collection_name/collection_names paths in the get_sources_from_items function perform vector store queries without any authorization check, allowing users to extract content from files and knowledge bases they do not have access to. This vulnerability is fixed in 0.9.0.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, FolderForm uses model_config = ConfigDict(extra='allow'), which permits arbitrary fields to pass through Pydantic validation and be included in model_dump(exclude_unset=True). In insert_new_folder, the server-assigned user_id is placed at the start of the dict and then overwritten by the spread of form data. Because FolderModel declares user_id: str as a real field (not just a form extra), any attacker-supplied user_id in the POST body is accepted by the model and persisted on the Folder row. This vulnerability is fixed in 0.9.0.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the /responses endpoint in the OpenAI router accepts any authenticated user and forwards requests directly to upstream LLM providers without enforcing per-model access control. While the primary chat completion endpoint (generate_chat_completion) checks model ownership, group membership, and AccessGrants before allowing a request, the /responses proxy only validates that the user has a valid session via get_verified_user. This allows any authenticated user to interact with any model configured on the instance by sending a POST request to /api/openai/responses with an arbitrary model ID. This vulnerability is fixed in 0.9.0.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the POST /api/v1/models/import endpoint allows users with the workspace.models_import permission to overwrite any existing model in the database, regardless of ownership. When an imported model's ID matches an existing model, the endpoint merges the attacker's payload over the existing model data and writes it to the database with no ownership or access grant validation. Additionally, filter_allowed_access_grants is never called, bypassing the access grant restrictions enforced on all other model mutation endpoints. This vulnerability is fixed in 0.9.0.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.6.19, there's an IDOR in the channels message management system that allows authenticated users to modify or delete any message within channels they have read access to. The vulnerability exists in the message update and delete endpoints, which implement channel-level authorization but completely lack message ownership validation. While the frontend correctly implements ownership checks (showing edit/delete buttons only for message owners or admins), the backend APIs bypass these protections by only validating channel access permissions without verifying that the requesting user owns the target message. This creates a client-side security control bypass where attackers can directly call the APIs to modify other users' messages. This vulnerability is fixed in 0.6.19.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.8.0, GET /api/v1/memories/ef is accessible without authentication and executes request.app.state.EMBEDDING_FUNCTION(...). This allows any unauthenticated caller to trigger embedding generation which can lead to direct cost exposure if a paid provider is used. This vulnerability is fixed in 0.8.0.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.8.6, in standard channels (i.e., channels whose channel.type is neither group nor dm), the endpoint POST /api/v1/channels/{channel_id}/messages/{message_id}/update can be accessed with read permission only. When access_control is set to None, the authorization check has_access(..., type="read") evaluates to True, allowing users who are not the message owner to update messages. As a result, unauthorized modification of other users’ messages is possible. This vulnerability is fixed in 0.8.6.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.8.6, there is a vulnerability in chat completion API, which allows attackers to bypass tool restrictions, potentially enabling unauthorized actions or access. In the chat_completion API, the parameters tool_ids and tool_servers are supplied by the user. These parameters are used to create a tools_dict by the middleware. This is then used by get_tool_by_id to retrieve the appropriate tool. However, there is no checks in that ensures the user that uses the API has permission to use the tool, meaning that a user can invoke any server tool by supplying the correct tool_id or tool_servers parameters via the chat completion API. Moreover, the authentication token stored in the server would be used when invoking the tool, so the tool will be invoked with the server privilege. This vulnerability is fixed in 0.8.6.

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to version 0.8.6, an access control check is missing when deleting a file from a knowledge base. The only check being done is that the user has write access to the knowledge base (or is admin), but NOT that the file actually belongs to this knowledge base. It is thus possible to delete arbitrary files from arbitrary knowledge bases (as long as one knows the file id). Version 0.8.6 patches the issue.

In version v0.3.8 of open-webui/open-webui, improper access control vulnerabilities allow an attacker to view any prompts. The application does not verify whether the attacker is an administrator, allowing the attacker to directly call the /api/v1/prompts/ interface to retrieve all prompt information created by the admin, which includes the ID values. Subsequently, the attacker can exploit the /api/v1/prompts/command/{command_id} interface to obtain arbitrary prompt information.

A privilege escalation vulnerability in Fortinet FortiOS 6.0.0 to 6.0.6, 5.6.0 to 5.6.10, 5.4 and below allows admin users to elevate their profile to super_admin via restoring modified configurations.

The Multiple Roles per User plugin for WordPress is vulnerable to unauthorized modification of data due to a missing capability check on the 'mrpu_add_multiple_roles_ui' and 'mrpu_save_multiple_user_roles' functions in all versions up to, and including, 1.0. This makes it possible for authenticated attackers, granted the 'edit_users' capability, to edit any user's role, including promoting users to Administrator and demoting Administrators to lower-privileged roles.

The ExactMetrics – Google Analytics Dashboard for WordPress (Website Stats Plugin) plugin for WordPress is vulnerable to unauthorized arbitrary plugin installation and activation in all versions up to, and including, 9.1.2. This is due to the reports page exposing the 'onboarding_key' transient to any user with the 'exactmetrics_view_dashboard' capability. This key is the sole authorization gate for the '/wp-json/exactmetrics/v1/onboarding/connect-url' REST endpoint, which returns a one-time hash (OTH) token. This OTH token is then the only credential checked by the 'exactmetrics_connect_process' AJAX endpoint — which has no capability check, no nonce verification, and accepts an arbitrary plugin ZIP URL via the file parameter for installation and activation. This makes it possible for authenticated attackers, with Editor-level access and above granted the report viewing permission, to install and activate arbitrary plugins from attacker-controlled URLs, leading to Remote Code Execution.

An issue has been discovered in GitLab EE affecting all versions from 16.6 before 18.2.7, 18.3 before 18.3.3, and 18.4 before 18.4.1. Project Maintainers can exploit a vulnerability where they can assign custom roles to users with permissions exceeding their own, effectively granting themselves elevated privileges.

Ella Core is a 5G core designed for private networks. Prior to version 1.7.0, the NetworkManager role was granted backup and restore permission. The restore endpoint accepted any valid SQLite file without verifying its contents. A NetworkManager could replace the production database with a tampered copy to escalate to Admin, gaining access to user management, audit logs, debug endpoints, and operator identity configuration that the role was explicitly denied. In version 1.7.0, backup and restore permissions have been removed from the NetworkManager role.

Ghost is a Node.js content management system. An error in the implementation of the limits service between versions 4.0.0 and 4.9.4 allows all authenticated users (including contributors) to view admin-level API keys via the integrations API endpoint, leading to a privilege escalation vulnerability. This issue is patched in Ghost version 4.10.0. As a workaround, disable all non-Administrator accounts to prevent API access. It is highly recommended to regenerate all API keys after patching or applying the workaround.

The Uncanny Groups for LearnDash plugin for WordPress is vulnerable to privilege escalation in all versions up to, and including, 6.1.0.1. This is due to the plugin not properly restricting what users a group leader can edit. This makes it possible for authenticated attackers, with group leader-level access and above, to change admin account email addresses which can subsequently lead to admin account access.

A Improper Privilege Management vulnerability in SUSE Rancher allows users with the restricted-admin role to escalate to full admin. This issue affects: SUSE Rancher Rancher versions prior to 2.5.13; Rancher versions prior to 2.6.4.

Contao is an open source CMS that allows creation of websites and scalable web applications. In affected versions it is possible to gain privileged rights in the Contao back end. Installations are only affected if they have untrusted back end users who have access to the form generator. All users are advised to update to Contao 4.4.56, 4.9.18 or 4.11.7. As a workaround users may disable the form generator or disable the login for untrusted back end users.

The JetFormBuilder plugin for WordPress is vulnerable to privilege escalation in all versions up to, and including, 3.3.4.1. This is due to improper restriction on user meta fields. This makes it possible for authenticated attackers, with administrator-level and above permissions, to register as super-admins on the sites configured as multi-sites.

StudioCMS is a server-side-rendered, Astro native, headless content management system. Prior to 0.4.3, the REST API createUser endpoint uses string-based rank checks that only block creating owner accounts, while the Dashboard API uses indexOf-based rank comparison that prevents creating users at or above your own rank. This inconsistency allows an admin to create additional admin accounts via the REST API, enabling privilege proliferation and persistence. This vulnerability is fixed in 0.4.3.

Umbraco is an ASP.NET CMS. From 15.3.1 to before 16.5.1 and 17.2.2, A privilege escalation vulnerability has been identified in Umbraco CMS. Under certain conditions, authenticated backoffice users with permission to manage users, may be able to elevate their privileges due to insufficient authorization enforcement when modifying user group memberships. The affected functionality does not properly validate whether a user has sufficient privileges to assign highly privileged roles. This vulnerability is fixed in 16.5.1 and 17.2.2.

Missing Authorization vulnerability in Ninja Team Filebird filebird allows Exploiting Incorrectly Configured Access Control Security Levels.This issue affects Filebird: from n/a through <= 6.3.2.

The BackWPup – WordPress Backup & Restore Plugin plugin for WordPress is vulnerable to unauthorized modification of data that can lead to privilege escalation due to a missing capability check on the save_site_option() function in all versions up to, and including, 5.6.2. This makes it possible for authenticated attackers, with level access and above, to update arbitrary options on the WordPress site. This can be leveraged to update the default role for registration to administrator and enable user registration for attackers to gain administrative user access to a vulnerable site.

Improper Privilege Management vulnerability in realmag777 HUSKY allows Privilege Escalation.This issue affects HUSKY: from n/a through 1.3.6.1.

An issue in Ocuco Innovation - JOBMANAGER.EXE v2.10.24.16 allows attackers to bypass authentication and escalate privileges to Administrator via a crafted TCP packet.

Improper Privilege Management vulnerability in WebAppick CTX Feed allows Privilege Escalation.This issue affects CTX Feed: from n/a through 6.5.6.

A command injection issue was discovered on Supermicro X11SSM-F, X11SAE-F, and X11SSE-F 1.66 devices. An attacker can exploit this to elevate privileges from a user with BMC administrative privileges.

Missing Authorization vulnerability in Muffingroup Betheme.This issue affects Betheme: from n/a through 27.1.1.

In MongoDB Ops Manager v5.0 prior to 5.0.22 and v6.0 prior to 6.0.17 it is possible for an authenticated user with project owner or project user admin access to generate an API key with the privileges of org owner resulting in privilege escalation.

In PHOENIX CONTACTs WP 6xxx series web panels in versions prior to 4.0.10 the SNMP daemon is running with root privileges allowing a remote attacker with knowledge of the SNMPv2 r/w community string to execute system commands as root.

Improper Privilege Management vulnerability in Crocoblock JetFormBuilder allows Privilege Escalation.This issue affects JetFormBuilder: from n/a through 3.0.8.

Improper Privilege Management vulnerability in WebToffee WooCommerce PDF Invoices, Packing Slips, Delivery Notes and Shipping Labels allows Privilege Escalation.This issue affects WooCommerce PDF Invoices, Packing Slips, Delivery Notes and Shipping Labels: from n/a through 4.2.1.

Scylladb is a NoSQL data store using the seastar framework, compatible with Apache Cassandra. Authenticated users who are authorized to create tables in a keyspace can escalate their privileges to access a table in the same keyspace, even if they don't have permissions for that table. This issue has not yet been patched. A workaround to address this issue is to disable CREATE privileges on a keyspace, and create new tables on behalf of other users.

Grafana is an open-source platform for monitoring and observability. The vulnerability impacts Grafana instances with several organizations, and allows a user with Organization Admin permissions in one organization to change the permissions associated with Organization Viewer, Organization Editor and Organization Admin roles in all organizations. It also allows an Organization Admin to assign or revoke any permissions that they have to any user globally. This means that any Organization Admin can elevate their own permissions in any organization that they are already a member of, or elevate or restrict the permissions of any other user. The vulnerability does not allow a user to become a member of an organization that they are not already a member of, or to add any other users to an organization that the current user is not a member of.

Multiple vulnerabilities in the web-based management interface of certain Cisco Small Business 100, 300, and 500 Series Wireless Access Points could allow an authenticated, remote attacker to obtain sensitive information from or inject arbitrary commands on an affected device. For more information about these vulnerabilities, see the Details section of this advisory.

Multiple vulnerabilities in Cisco SD-WAN vManage Software could allow an unauthenticated, remote attacker to execute arbitrary code or gain access to sensitive information, or allow an authenticated, local attacker to gain escalated privileges or gain unauthorized access to the application. For more information about these vulnerabilities, see the Details section of this advisory.

A vulnerability has been identified when granting a create or * global role for a resource type of "namespaces"; no matter the API group, the subject will receive * permissions for core namespaces. This can lead to someone being capable of accessing, creating, updating, or deleting a namespace in the project.