Open WebUI: Cross-user model-list exposure via static cache key in get_all_models (aiocache key= vs key_builder= misuse)
Open WebUI is an extensible, feature-rich, and user-friendly self-hosted AI platform. From 0.6.27 before 0.10.0, get_all_models handlers in routers/openai.py and routers/ollama.py passed a lambda to aiocache key instead of key_builder, causing permission-filtered per-user model lists to share a static cache entry and exposing one user’s model list to another caller during the TTL window. This issue is fixed in version 0.10.0.
Open WebUI: Upload `metadata.knowledge_id` bypasses the knowledge-base write-access check (read-only users can add files to KB)
Open WebUI is an extensible, feature-rich, and user-friendly self-hosted AI platform. Prior to 0.10.0, the file upload path accepted metadata.knowledge_id and auto-linked uploaded files to a target knowledge base without applying the write-access check used by /api/v1/knowledge//file/add, allowing read-only knowledge-base users to add arbitrary files. This issue is fixed in version 0.10.0.
Open WebUI: Cross-user code-interpreter and tool execution via unvalidated Socket.IO event-caller session_id
Open WebUI is an extensible, feature-rich, and user-friendly self-hosted AI platform. Prior to 0.10.0, get_event_call delivered execute:python and execute:tool Socket.IO events to a client-supplied session_id after checking only that the session was connected, allowing authenticated users who learned another socket ID through ydoc:document:join to run code interpreter Python or tools in that user session. This issue is fixed in version 0.10.0.
Open WebUI: Realtime endpoints accept Redis-revoked JWTs after signout/backchannel logout
Open WebUI is an extensible, feature-rich, and user-friendly self-hosted AI platform. From 0.9.0 before 0.10.0 with Redis configured, Socket.IO connect, user-join, join-channels, join-note, and the terminal websocket first-message authentication used decode_token without the Redis-backed is_valid_token revocation check, allowing revoked JWTs to continue authenticating realtime connections. This issue is fixed in version 0.10.0.
Open WebUI: Unauthenticated WebSocket Access to Collaborative Document Handlers (ydoc:awareness:update, ydoc:document:leave)
Open WebUI is an extensible, feature-rich, and user-friendly self-hosted AI platform. From 0.6.16 before 0.10.0, the Socket.IO server is configured with always_connect=True. The ydoc:awareness:update and ydoc:document:leave Socket.IO handlers accepted collaborative-document events without requiring an authenticated user, allowing unauthorized manipulation of document collaboration state. This issue is fixed in version 0.10.0.
Open WebUI: ReDoS in skill-mention regexes causes whole-instance DoS on default config
Open WebUI is an extensible, feature-rich, and user-friendly self-hosted AI platform. From 0.9.2 before 0.10.0, the SKILL_MENTION_RE and strip_re regular expressions in backend/open_webui/utils/middleware.py parsed <$skillId|label> skill mentions with overlapping quantifiers, allowing an authenticated chat message containing <$ without a closing > to trigger quadratic backtracking and block the asyncio event loop. This issue is fixed in version 0.10.0.
Open WebUI: Scheduled automations continue after pending-user deactivation and stored model ACL revocation
Open WebUI is an extensible, feature-rich, and user-friendly self-hosted AI platform. From 0.9.0 before 0.10.0, execute_automation rehydrated automation owners without rechecking that they were still active or still had features.automations, and check_model_access only enforced private-model grants for the exact user role, allowing deactivated pending users to continue scheduled model execution. This issue is fixed in version 0.10.0.
Open WebUI: POST /api/v1/images/edit bypasses the global image-edit switch and the per-user image-generation permission
Open WebUI is an extensible, feature-rich, and user-friendly self-hosted AI platform. From 0.8.11 before 0.10.0, POST /api/v1/images/edit required only a verified account and did not enforce the global image-edit switch or the per-user image-generation permission, allowing a non-admin user to invoke server-side image editing with administrator-configured provider credentials. This issue is fixed in version 0.10.0.
Open WebUI: Account enumeration via observable login timing discrepancy
Open WebUI is an extensible, feature-rich, and user-friendly self-hosted AI platform. Prior to 0.10.0, the /api/v1/auths/signin endpoint looked users up by email and only ran bcrypt password verification when a credential existed, making registered-account attempts measurably slower than missing-email attempts and allowing unauthenticated account enumeration. This issue is fixed in version 0.10.0.
Open WebUI is an extensible, feature-rich, and user-friendly self-hosted AI platform. Prior to 0.10.0, Open WebUI runs client-side Python with Pyodide in a same-origin web worker, allowing stored chat payloads that use pyodide.http.pyfetch or the js module fetch and XMLHttpRequest APIs to issue authenticated same-origin requests when a victim clicks Run, which can reach admin-only endpoints and execute server-side code through configured tools. This issue is fixed in version 0.10.0.
Phantom: Arbitrary file write and decode-bomb DoS via unconfined MCP tool paths
### Impact
In Phantom <= 1.3.0, when `PHANTOM_OUTPUT_DIR` was unset (the default), the MCP tools accepted arbitrary absolute output paths with no confinement. Anything able to send tool calls (e.g. an AI agent driving the MCP interface) could **write or overwrite arbitrary files** the process user can write — including shell startup files (`~/.zshrc`) or a Reaper `__startup.lua`, which is effectively local code execution on a developer workstation.
Separately, the stem-separation and render paths decoded input audio with no size/duration cap (the analysis path was already guarded). A small, highly compressed FLAC/OGG could expand to multi-g
Stanza: Remote Code Execution via Unsafe Pickle Deserialization in Model Loaders
Stanza is a Stanford NLP Python library for tokenization, sentence segmentation, NER, and parsing of many human languages. Prior to 1.12.2, Stanza model loaders such as stanza.models.common.pretrain.Pretrain.load() attempt torch.load(..., weights_only=True) but fall back to torch.load(..., weights_only=False) on attacker-controllable pickle.UnpicklingError, allowing a malicious .pt pretrain or model file to execute arbitrary pickle code when a Stanza NLP pipeline loads it. This issue is fixed in version 1.12.2.
RestrictedPython guard hooks can be shadowed via positional-only arguments
RestrictedPython is a tool that helps to define a subset of the Python language which allows to provide a program input into a trusted environment. Prior to 8.3, check_function_argument_names() rejected protected guard hook names for regular, variadic, and keyword-only arguments but omitted positional-only arguments, allowing __getattr__, _getitem_, _write_, or _print_ to be shadowed by a local parameter and bypass the embedding application's access policy. This issue is fixed in version 8.3.
OpenCTI: Elasticsearch Painless Script Injection via GraphQL `script` filter operator allows authenticated user to exfiltrate data and cause DoS
OpenCTI is an open source platform for managing cyber threat intelligence knowledge and observables. Prior to 7.260401.0, the OpenCTI GraphQL API exposes a script filter operator in its FilterOperator enum that allows any authenticated user with the KNOWLEDGE capability to pass user-supplied Elasticsearch Painless script values directly into search queries without validation or sanitization, allowing computationally expensive scripts to consume cluster CPU resources and degrade or deny service for all users. This issue is fixed in version 7.260401.0.
OpenCTI: Authorization Bypass via `synchronized-upsert` HTTP Header Injection
OpenCTI is an open source platform for managing cyber threat intelligence knowledge and observables. Prior to 7.260326.0, an authorization bypass vulnerability in OpenCTI allows any authenticated user with KNOWLEDGE_KNUPDATE permission to bypass Confidence Level validation and Object Marking restrictions by injecting the synchronized-upsert: true HTTP header, enabling attackers to downgrade confidence levels, remove security markings such as TLP:RED, manipulate relationships, and affect STIX object types including Indicators, ThreatActors, Malware, and Reports. This issue is fixed in version 7.260326.0.
JupyterLab Git is a Git extension for JupyterLab. Prior to 0.54.0, jupyterlab-git uses fnmatch.fnmatchcase() in GitHandler.prepare() in jupyterlab_git/handlers.py to enforce excluded_paths, allowing an authenticated user on a case-insensitive filesystem to vary URL path casing and read excluded directories. This issue is fixed in version 0.54.0.
JupyterLab Git is a Git extension for JupyterLab. From 0.30.0b3 before 0.54.0, the PlainTextDiff.ts createHeader() method passes Git filenames directly to innerHTML when rendering renamed files in commit history, allowing a crafted filename to execute JavaScript when a victim views the rename diff in the Git History tab. This issue is fixed in version 0.54.0.
AsyncSSH AuthorizedKeysFile username substitution bypass through ~ and environment expansion
AsyncSSH is a Python package which provides an asynchronous client and server implementation of the SSHv2 protocol on top of the Python asyncio framework. Version 2.23.0 contains an incomplete fix for CVE-2026-45309 in SSHServerConfig._set_tokens that blocks /, , and .. before %u substitution in AuthorizedKeysFile but does not block a leading ~ or ${ENV}, allowing later expansion in _expand_val and Path(filename).expanduser() to escape the intended authorized-keys directory. This issue is fixed in version 2.23.1.
AsyncSSH: SCP Path Traversal to Arbitrary File Write
AsyncSSH is a Python package which provides an asynchronous client and server implementation of the SSHv2 protocol on top of the Python asyncio framework. Prior to 2.23.1, a malicious SSH server can write arbitrary files on the asyncssh SCP client's filesystem by sending filenames containing ../ traversal sequences because _parse_cd_args in scp.py returns server-provided names verbatim and _recv_files joins them to the destination path without enforcing the target directory boundary. This issue is fixed in version 2.23.1.
py7zr: O(n^2) algorithmic complexity DoS in PackInfo._read()
py7zr is a Python-based library and utility to support 7zip archive compression, decompression, encryption and decryption. Prior to 1.1.3, PackInfo._read() in archiveinfo.py used an O(n^2) cumulative sum pattern for attacker-controlled numstreams values parsed from archive headers, allowing a crafted .7z archive to cause excessive CPU consumption during SevenZipFile.init() before extraction. This issue is fixed in version 1.1.3.