Search
CVE Explorer
Search the full tracked CVE corpus across every vendor — by keyword, vendor, severity, CVSS band and publication date. Server-rendered; each filtered view has its own URL.
01
Filters
Submit to refine — state is held in the URL.
02
Results
43,573 matching · page 273/872Each CVE id links to its NVD record.
| CVE | Severity | CVSS | Summary | Published |
|---|---|---|---|---|
| CVE-2018-25341(opens NVD record) | High | 8.2 | Smartshop 1 contains a SQL injection vulnerability that allows unauthenticated attackers to execute arbitrary SQL queries by injecting malicious code through the id parameter. Attackers can send GET requests to product.php with union-based SQL injection payloads in the id parameter to extract sensitive database information including usernames and database names. | May 23, 2026 |
| CVE-2018-25340(opens NVD record) | High | 8.2 | Smartshop 1 contains a SQL injection vulnerability that allows unauthenticated attackers to execute arbitrary SQL queries by injecting malicious code through the id parameter. Attackers can send GET requests to category.php with UNION-based SQL injection payloads in the id parameter to extract sensitive database information including usernames and other data. | May 23, 2026 |
| CVE-2026-9306(opens NVD record) | Low | 3.7 | A security vulnerability has been detected in QuantumNous new-api up to 0.12.1. This affects the function RelayMidjourneyImage/GetByOnlyMJId of the file router/relay-router.go of the component Midjourney Image Relay Endpoint. Such manipulation leads to authorization bypass. The attack can be launched remotely. The attack requires a high level of complexity. The exploitability is reported as difficult. The exploit has been disclosed publicly and may be used. The vendor was contacted early about this disclosure but did not respond in any way. | May 23, 2026 |
| CVE-2026-9305(opens NVD record) | Medium | 6.3 | A weakness has been identified in QuantumNous new-api up to 0.12.1. The impacted element is the function SearchUserTopUps/SearchAllTopUps of the file model/topup.go of the component self Endpoint. This manipulation causes sql injection. The attack can be initiated remotely. The exploit has been made available to the public and could be used for attacks. The vendor was contacted early about this disclosure but did not respond in any way. | May 23, 2026 |
| CVE-2026-9304(opens NVD record) | Medium | 5.0 | A security flaw has been discovered in calcom cal.diy up to 4.9.4. The affected element is the function validateUrlForSSRF of the file apps/web/app/api/logo/route.ts of the component Logo API. The manipulation results in server-side request forgery. It is possible to launch the attack remotely. Attacks of this nature are highly complex. The exploitability is described as difficult. The exploit has been released to the public and may be used for attacks. The vendor was contacted early about this disclosure but did not respond in any way. | May 23, 2026 |
| CVE-2026-9303(opens NVD record) | Medium | 4.3 | A vulnerability was identified in calcom cal.diy up to 4.9.4. Impacted is an unknown function. The manipulation leads to cross-site request forgery. It is possible to initiate the attack remotely. The exploit is publicly available and might be used. The vendor was contacted early about this disclosure but did not respond in any way. | May 23, 2026 |
| CVE-2026-9302(opens NVD record) | Medium | 6.3 | A vulnerability was determined in 546669204 vps-inventory-monitoring up to 98c00b370668c96ae75e91c15548d9ea113652d9. This issue affects the function eval of the file app/index/command/VpsTest.php of the component VpsTest Console. Executing a manipulation of the argument vf can lead to code injection. The attack may be performed from remote. The exploit has been publicly disclosed and may be utilized. This product utilizes a rolling release system for continuous delivery, and as such, version information for affected or updated releases is not disclosed. The project was informed of the problem early through an issue report but has not responded yet. | May 23, 2026 |
| CVE-2026-9301(opens NVD record) | Medium | 6.3 | A vulnerability was found in omec-project amf up to 2.1.1. This vulnerability affects unknown code of the component NGReset Message Handler. Performing a manipulation results in memory corruption. The attack is possible to be carried out remotely. The exploit has been made public and could be used. It is recommended to apply a patch to fix this issue. | May 23, 2026 |
| CVE-2026-9300(opens NVD record) | Medium | 6.3 | A vulnerability has been found in omec-project amf up to 2.1.1. This affects an unknown part of the component NGSetupRequest Handler. Such manipulation leads to memory corruption. The attack can be executed remotely. The exploit has been disclosed to the public and may be used. It is best practice to apply a patch to resolve this issue. | May 23, 2026 |
| CVE-2026-46300(opens NVD record) | High | 7.8 | In the Linux kernel, the following vulnerability has been resolved: net: skbuff: preserve shared-frag marker during coalescing skb_try_coalesce() can attach paged frags from @from to @to. If @from has SKBFL_SHARED_FRAG set, the resulting @to skb can contain the same externally-owned or page-cache-backed frags, but the shared-frag marker is currently lost. That breaks the invariant relied on by later in-place writers. In particular, ESP input checks skb_has_shared_frag() before deciding whether an uncloned nonlinear skb can skip skb_cow_data(). If TCP receive coalescing has moved shared frags into an unmarked skb, ESP can see skb_has_shared_frag() as false and decrypt in place over page-cache backed frags. Propagate SKBFL_SHARED_FRAG when skb_try_coalesce() transfers paged frags. The tailroom copy path does not need the marker because it copies bytes into @to's linear data rather than transferring frag descriptors. | May 23, 2026 |
| CVE-2026-43503(opens NVD record) | High | 8.8 | In the Linux kernel, the following vulnerability has been resolved: net: skbuff: propagate shared-frag marker through frag-transfer helpers Two frag-transfer helpers (__pskb_copy_fclone() and skb_shift()) fail to propagate the SKBFL_SHARED_FRAG bit in skb_shinfo()->flags when moving frags from source to destination. __pskb_copy_fclone() defers the rest of the shinfo metadata to skb_copy_header() after copying frag descriptors, but that helper only carries over gso_{size,segs, type} and never touches skb_shinfo()->flags; skb_shift() moves frag descriptors directly and leaves flags untouched. As a result, the destination skb keeps a reference to the same externally-owned or page-cache-backed pages while reporting skb_has_shared_frag() as false. The mismatch is harmful in any in-place writer that uses skb_has_shared_frag() to decide whether shared pages must be detoured through skb_cow_data(). ESP input is one such writer (esp4.c, esp6.c), and a single nft 'dup to <local>' rule -- or any other nf_dup_ipv4() / xt_TEE caller -- is enough to land a pskb_copy()'d skb in esp_input() with the marker stripped, letting an unprivileged user write into the page cache of a root-owned read-only file via authencesn-ESN stray writes. Set SKBFL_SHARED_FRAG on the destination whenever frag descriptors were actually moved from the source. skb_copy() and skb_copy_expand() share skb_copy_header() too but linearize all paged data into freshly allocated head storage and emerge with nr_frags == 0, so skb_has_shared_frag() returns false on its own; they need no change. The same omission exists in skb_gro_receive() and skb_gro_receive_list(). The former moves the incoming skb's frag descriptors into the accumulator's last sub-skb via two paths (a direct frag-move loop and the head_frag + memcpy path); the latter chains the incoming skb whole onto p's frag_list. Downstream skb_segment() reads only skb_shinfo(p)->flags, and skb_segment_list() reuses each sub-skb's shinfo as the nskb -- both p and lp must carry the marker. The same omission also exists in tcp_clone_payload(), which builds an MTU probe skb by moving frag descriptors from skbs on sk_write_queue into a freshly allocated nskb. The helper falls into the same family and warrants the same fix for consistency; no TCP TX-side in-place writer is currently known to reach a user page through this gap, but a future consumer depending on the marker would regress silently. The same omission exists in skb_segment(): the per-iteration flag merge takes only head_skb's flag, and the inner switch that rebinds frag_skb to list_skb on head_skb-frags exhaustion does not fold the new frag_skb's flag into nskb. Fold frag_skb's flag at both sites so segments drawing frags from frag_list members carry the marker. | May 23, 2026 |
| CVE-2026-9299(opens NVD record) | Medium | 6.3 | A flaw has been found in omec-project amf up to 2.1.1. Affected by this issue is the function PDUSessionResourceModifyIndication of the file /go/src/amf/ngap/handler.go. This manipulation causes memory corruption. Remote exploitation of the attack is possible. The exploit has been published and may be used. Applying a patch is the recommended action to fix this issue. | May 23, 2026 |
| CVE-2026-9298(opens NVD record) | Medium | 6.3 | A vulnerability was detected in omec-project amf up to 2.1.1. Affected by this vulnerability is an unknown functionality of the component PathSwitchRequest Handler. The manipulation results in memory corruption. The attack may be launched remotely. The exploit is now public and may be used. It is advisable to implement a patch to correct this issue. | May 23, 2026 |
| CVE-2026-9297(opens NVD record) | Medium | 6.3 | A security vulnerability has been detected in Edimax BR-6428NS 1.10. Affected is the function formWlbasic of the file /goform/formWlbasic of the component POST Request Handler. The manipulation of the argument repeaterSSID leads to command injection. The attack may be initiated remotely. The exploit has been disclosed publicly and may be used. The vendor was contacted early about this disclosure but did not respond in any way. | May 23, 2026 |
| CVE-2026-9296(opens NVD record) | Medium | 6.3 | A weakness has been identified in Edimax BR-6428NS 1.10. This impacts the function system of the file /goform/formWlanM of the component POST Request Handler. Executing a manipulation of the argument ateFunc/ateGain/ateTxCount/ateChan/ateRate/ateMacID/e2pTxPower1/e2pTxPower2/e2pTxPower3/e2pTxPower4/e2pTxPower5/e2pTxPower6/e2pTxPower7/e2pTx2Power1/e2pTx2Power2/e2pTx2Power3/e2pTx2Power4/e2pTx2Power5/e2pTx2Power6/e2pTx2Power7/ateTxFreqOffset/ateMode/ateBW/ateAntenna/e2pTxFreqOffset/e2pTxPwDeltaB/e2pTxPwDeltaG/e2pTxPwDeltaMix/e2pTxPwDeltaN/readE2P can lead to command injection. The attack can be launched remotely. The exploit has been made available to the public and could be used for attacks. The vendor was contacted early about this disclosure but did not respond in any way. | May 23, 2026 |
| CVE-2026-9295(opens NVD record) | High | 8.8 | A security flaw has been discovered in Edimax BR-6428NS 1.10. This affects the function formWirelessTbl of the file /goform/formWirelessTbl of the component POST Request Handler. Performing a manipulation of the argument vapurl results in buffer overflow. The attack can be initiated remotely. The exploit has been released to the public and may be used for attacks. The vendor was contacted early about this disclosure but did not respond in any way. | May 23, 2026 |
| CVE-2026-9294(opens NVD record) | High | 8.8 | A vulnerability was identified in Edimax BR-6428NS 1.10. The impacted element is the function formWanTcpipSetup of the file /goform/formWanTcpipSetup of the component POST Request Handler. Such manipulation of the argument pppUserName leads to buffer overflow. It is possible to launch the attack remotely. The exploit is publicly available and might be used. The vendor was contacted early about this disclosure but did not respond in any way. | May 23, 2026 |
| CVE-2026-9284(opens NVD record) | High | 8.2 | The WooCommerce PayPal Payments plugin for WordPress is vulnerable to unauthorized order manipulation and information disclosure due to missing authorization checks on the `ppc-create-order` and `ppc-get-order` WC-AJAX endpoints in all versions up to, and including, 4.0.1. The `ppc-create-order` endpoint accepts an arbitrary WooCommerce order ID in the `pay-now` context without validating order ownership, allowing attackers to create PayPal orders for any WC order and write PayPal metadata to it. The `ppc-get-order` endpoint returns full PayPal order details for any PayPal order ID without binding to the requester's session. This makes it possible for unauthenticated attackers to chain these endpoints to manipulate other customers' order payment flows and exfiltrate sensitive order details (payer information, shipping data) by creating a PayPal order for a victim's WC order and then retrieving the PayPal order data. | May 23, 2026 |
| CVE-2026-6898(opens NVD record) | High | 8.8 | The Wishlist Member plugin for WordPress is vulnerable to unauthorized modification of data due to a missing capability check on the 'WishListMember3_Hooks::generate_api_key' function in all versions up to, and including, 3.30.1. This makes it possible for authenticated attackers, with Subscriber-level access and above, to update the REST API Secret Key, which can be used to create a new membership level assigned the administrator WordPress role, and register an arbitrary administrator-level user account, resulting in complete site takeover. | May 23, 2026 |
| CVE-2026-6897(opens NVD record) | High | 8.8 | The Wishlist Member plugin for WordPress is vulnerable to unauthorized modification of data due to a missing capability check on the 'WishListMember\Features\Team_Accounts::save_settings' function in all versions up to, and including, 3.30.1. This makes it possible for authenticated attackers, with Subscriber-level access and above, to update arbitrary plugin options, includes the REST API Secret Key, which can be used to create a new membership level assigned the administrator WordPress role, and register an arbitrary administrator-level user account, resulting in complete site takeover. | May 23, 2026 |
| CVE-2026-6895(opens NVD record) | High | 8.8 | The WishList Member plugin for WordPress is vulnerable to Missing Authorization leading to Sensitive Information Disclosure and Privilege Escalation in versions up to and including 3.30.1. This is due to the missing capability checks in the 'export_settings' function. This function returns the REST API Secret Key to the attacker in the AJAX JSON response. An attacker who obtains this key can authenticate to the WishList Member API, create a new membership level assigned the administrator WordPress role, and register an arbitrary administrator-level user account, resulting in complete site takeover. | May 23, 2026 |
| CVE-2026-6419(opens NVD record) | High | 8.8 | The WishList Member plugin for WordPress is vulnerable to Privilege Escalation via Missing Authorization in versions up to and including 3.30.1. This is due to the missing capability and nonce check in the ajax_get_screen() function. This makes it possible for authenticated attackers, with Subscriber-level access and above, to supply an arbitrary admin screen identifier via the data[url] parameter, causing the plugin to load and execute the administrative API configuration template without authorization. The rendered HTML, which contains the plugin's plaintext REST API Secret Key, is returned directly to the attacker in the AJAX JSON response. An attacker who obtains this key can authenticate to the WishList Member API, create a new membership level assigned the administrator WordPress role, and register an arbitrary administrator-level user account, resulting in complete site takeover. | May 23, 2026 |
| CVE-2026-47280(opens NVD record) | Critical | 10.0 | Improper authentication in Azure Resource Manager (ARM) allows an unauthorized attacker to elevate privileges over a network. | May 22, 2026 |
| CVE-2026-45659(opens NVD record) | High | 8.8 | Deserialization of untrusted data in Microsoft Office SharePoint allows an authorized attacker to execute code over a network. | May 22, 2026 |
| CVE-2026-42901(opens NVD record) | Critical | 10.0 | Origin validation error in Microsoft Entra ID allows an unauthorized attacker to elevate privileges over a network. | May 22, 2026 |
| CVE-2026-42827(opens NVD record) | Medium | 6.5 | Improper neutralization of special elements used in a command ('command injection') in M365 Copilot allows an unauthorized attacker to disclose information over a network. | May 22, 2026 |
| CVE-2026-41149(opens NVD record) | Unscored | — | Mermaid is a JavaScript tool that uses Markdown-inspired text to create and modify diagrams and charts. Versions 10.9.5 and earlier, as well as 11.0.0-alpha.1 through 11.14.0, are vulnerable to HTML injection under the default configuration. Specifically, the classDef directive in Mermaid state diagrams permits DOM injection that escapes the SVG context. However, <script> tags are stripped, which prevents cross-site scripting (XSS). This issue has been fixed in versions 10.9.6 and 11.15.0. If developers are unable to immediately upgrade, they can work around this issue by setting "securityLevel": "sandbox", which prevents the issue by rendering the mermaid diagram in a sandboxed <iframe>. | May 22, 2026 |
| CVE-2026-41148(opens NVD record) | Unscored | — | Mermaid is a JavaScript tool that uses Markdown-inspired text to create and modify diagrams and charts. Versions 10.9.5 and prior, in addition to 11.0.0-alpha.1 through 11.12.0 are vulnerable to CSS injection through improper sanitization. The state diagram (and any other diagram type that routes user-controlled style strings through the createCssStyles parser) captures classDef values using an unrestricted regex that matches everything up to a newline. That value then flows unsanitized through addStyleClass() into createCssStyles() and is assigned to style.innerHTML, so a closing brace (}) in the value terminates the generated CSS selector and turns everything after it into a new CSS rule on the page. This enables page defacement, user tracking via url() callbacks, and DOM attribute exfiltration. This issue has been fixed in versions 10.9.6 and 11.15.0. If developers are unable to immediately upgrade, they can work around this issue by setting "securityLevel": "sandbox", which prevents the issue by rendering the mermaid diagram in a sandboxed <iframe>. | May 22, 2026 |
| CVE-2026-41104(opens NVD record) | Critical | 10.0 | Deserialization of untrusted data in Microsoft Planetary Computer Pro allows an unauthorized attacker to disclose information over a network. | May 22, 2026 |
| CVE-2026-41090(opens NVD record) | Critical | 9.3 | Improper neutralization of special elements used in a command ('command injection') in Microsoft Copilot allows an unauthorized attacker to perform tampering over a network. | May 22, 2026 |
| CVE-2026-40412(opens NVD record) | Critical | 10.0 | Unrestricted upload of file with dangerous type in Azure Orbital Spatio allows an unauthorized attacker to execute code over a network. | May 22, 2026 |
| CVE-2026-40411(opens NVD record) | Critical | 9.9 | Improper input validation in Azure Virtual Network Gateway allows an authorized attacker to execute code over a network. | May 22, 2026 |
| CVE-2026-35430(opens NVD record) | High | 8.8 | Authorization bypass through user-controlled key in Azure Privileged Identity Management (PIM) allows an authorized attacker to elevate privileges over a network. | May 22, 2026 |
| CVE-2026-33843(opens NVD record) | Critical | 9.1 | Authentication bypass using an alternate path or channel in Microsoft Azure Active Directory B2C allows an unauthorized attacker to elevate privileges over a network. | May 22, 2026 |
| CVE-2026-26147(opens NVD record) | High | 7.7 | Improper input validation in Azure Compute Gallery allows an authorized attacker to disclose information over a network. | May 22, 2026 |
| CVE-2026-23663(opens NVD record) | High | 7.5 | Improper privilege management in Azure Entra ID allows an unauthorized attacker to elevate privileges over a network. | May 22, 2026 |
| CVE-2026-23652(opens NVD record) | Critical | 10.0 | Improper neutralization of special elements used in a command ('command injection') in Microsoft Power Pages allows an unauthorized attacker to execute code over a network. | May 22, 2026 |
| CVE-2026-41147(opens NVD record) | High | 8.7 | NukeViet CMS is a multi Content Management System. Versions 4.5.07 and prior contain a Stored Cross-Site Scripting (XSS) vulnerability caused by insufficient server-side input sanitization in the Request class. The application relies primarily on client-side filtering to sanitize HTML tags and attributes in user-submitted content, which can be bypassed by intercepting and modifying HTTP requests directly (e.g., using Burp Suite). An attacker can inject malicious payloads which are stored server-side and executed in the browser of any user who views the content. Anyone viewing user-submitted content (such as administrators and moderators reviewing contact messages or comments) is impacted, and the vulnerability can be exploited by any anonymous visitor without authentication, with the Contact module used only as a proof of concept. Potential consequences include session hijacking through cookie theft, unauthorized actions performed under the victim's identity, defacement or redirection to phishing pages, and phishing attacks via manipulated email notifications. This issue has been fixed in version 4.5.08. If developers are unable to upgrade immediately, they should work around this issue by implementing server-side HTML sanitization in the Request class to strip or encode dangerous tags and attributes (e.g., <iframe>, srcdoc, event handlers like onerror/onload), enforcing a Content Security Policy (CSP) to restrict inline script execution, and set cookies with the HttpOnly flag to mitigate cookie theft via XSS. | May 22, 2026 |
| CVE-2026-41076(opens NVD record) | High | 8.1 | RT is an open source, enterprise-grade issue and ticket tracking system. Versions 5.0.9 and prior in addition to 6.0.0 through 6.0.2 contain an authentication bypass vulnerability in RT installations that use LDAP/AD for user authentication. Under certain LDAP server configurations, an attacker may be able to authenticate as any LDAP-backed RT user without supplying valid credentials. This issue has been fixed in versions 5.0.10 and 6.0.3. If developers are unable to upgrade immediately, they can temporarily work around this issue by reviewing their LDAP server's authentication policy to ensure it rejects unauthenticated bind attempts. Upgrading RT remains the recommended fix. | May 22, 2026 |
| CVE-2026-41075(opens NVD record) | High | 8.8 | RT is an open source, enterprise-grade issue and ticket tracking system. Versions 5.0.0 through 5.0.9 and 6.0.0 through 6.0.2 contain an SQL injection vulnerability. An authenticated user can craft input that is incorporated into database queries without proper validation, potentially allowing them to read or modify data in the RT database. This issue has been fixed in versions 5.0.10 and 6.0.3. If developers are unable to upgrade immediately, they can temporarily work around this issue by restricting RT account access to trusted users. | May 22, 2026 |
| CVE-2026-41074(opens NVD record) | High | 7.1 | RT is an open source, enterprise-grade issue and ticket tracking system. Versions 6.0.0 through 6.0.2 contain a Cross-Site Request Forgery (CSRF) vulnerability. An attacker who can induce a logged-in RT user to visit a malicious web page can trigger arbitrary state-changing actions in RT on that user's behalf. This issue has been fixed in version 6.0.3. | May 22, 2026 |
| CVE-2026-41073(opens NVD record) | Medium | 4.6 | RT is an open source, enterprise-grade issue and ticket tracking system. Versions prior to 5.0.10 and 6.0.0 through 6.0.2 contain a spreadsheet (CSV/formula) injection vulnerability. User-controlled data in spreadsheet exports is not sanitized before being written to the output file, which can cause spreadsheet applications to interpret crafted values as formulas or macros when the file is opened. This issue has been fixed in versions 5.0.10 and 6.0.3. If developers are unable to upgrade immediately, they can temporarily work around this issue by avoiding opening exported RT spreadsheet files directly in spreadsheet applications when the data may contain untrusted user input. | May 22, 2026 |
| CVE-2026-41071(opens NVD record) | High | 8.1 | libheif is a HEIF and AVIF file format decoder and encoder. In versions 1.21.2 and prior, a crafted HEIF sequence file where the saiz box declares more samples than actually exist in the track's chunk table causes a heap-buffer-overflow (out-of-bounds read) in the SampleAuxInfoReader constructor. The SampleAuxInfoReader constructor iterates over saiz->get_num_samples() samples but doesn't validate that this count is consistent with the number of chunks in the chunks vector. When saiz declares more samples than the chunks cover, the loop increments current_chunk past chunks.size(), causing an out-of-bounds read on the chunks vector. The vulnerability is triggered during file parsing (heif_context_read_from_file) without any additional user interaction. Any application using libheif to open untrusted HEIF files is affected. This issue has been fixed in version 1.22.0. | May 22, 2026 |
| CVE-2026-41069(opens NVD record) | Medium | 6.5 | libheif is a HEIF and AVIF file format decoder and encoder. In versions 1.21.2 and prior, a malformed HEIF sequence file can trigger an out-of-bounds read in core sequence parsing logic, causing DoS. A malformed file can have stco.entry_count == 0 (creating no chunks) while still passing validation because saio.entry_count == 0 matches, but with saiz.sample_count > 0 the SampleAuxInfoReader constructor still enters its loop. This leads to an out-of-bounds dereference on the empty chunks[0] in chunked mode. | May 22, 2026 |
| CVE-2026-40864(opens NVD record) | Medium | 5.4 | JupyterHub is software that allows users to create a multi-user server for Jupyter notebooks. In versions 4.1.0 through 5.4.4, XSRF protection (updated in 4.1.0) inappropriately treated requests with Sec-Fetch-Mode: no-cors as same-origin requests, bypassing XSRF checks. The JSON API is not affected, only HTTP form endpoints, such as /hub/spawn and /hub/accept-share, meaning attackers could trigger server spawn (but not access the server) and if the attacker is a JupyterHub user permitted to share access to their server, cause a user to accept a share and have access to the attacker's server. This issue has been fixed in version 5.4.5. If developers are unable to immediately upgrade, they can temporarily mitigate this issue by dropping requests to JupyterHub with Sec-Fetch-Mode: no-cors if they are using a reverse proxy. | May 22, 2026 |
| CVE-2026-3294(opens NVD record) | High | 8.8 | An authentication logic vulnerability in multiple TP-Link range extenders allows an unauthenticated attacker on an adjacent network to manipulate a login parameter and reset the administrator password due to insufficient validation. Successful exploitation allows an attacker to obtain full administrative control of the affected device, potentially impacting on confidentiality, integrity, and availability. | May 22, 2026 |
| CVE-2026-5843(opens NVD record) | High | 8.2 | The MLX inference backend in Docker Model Runner on macOS uses the MLX-LM library, which unconditionally imports and executes arbitrary Python files from model directories via the model_file configuration field in config.json. When a model's config.json specifies a model_file pointing to a Python file, MLX-LM uses importlib to load and execute it with no trust_remote_code gate or equivalent safety check. The MLX backend runs without sandboxing, resulting in arbitrary code execution on the Docker host as the Docker Desktop user. Any container on the Docker network can trigger this by calling the model-runner.docker.internal API to pull a malicious model from an attacker-controlled OCI registry and request inference. | May 22, 2026 |
| CVE-2026-5817(opens NVD record) | High | 8.2 | The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trust_remote_code=True when loading model tokenizers, and runs without sandboxing. This causes transformers.AutoTokenizer.from_pretrained() to import and execute arbitrary Python files included in any model pulled from an OCI registry, resulting in arbitrary code execution on the Docker host as the Docker Desktop user when inference is triggered. Any container on the Docker network can trigger this by calling the model-runner.docker.internal API to pull a malicious model and request inference. | May 22, 2026 |
| CVE-2026-40610(opens NVD record) | Medium | 5.5 | BentoML is a Python library for building online serving systems optimized for AI apps and model inference. In versions 1.4.38 and prior, the build packaging workflow follows attacker-controlled symlinks inside the build context and copies the referenced file contents into the generated Bento artifact. If a victim builds an untrusted repository or other attacker-supplied build context, the attacker can place a symlink such as loot.txt -> /tmp/outside-marker.txt or a link to a more sensitive local file. When bentoml build runs, BentoML dereferences the symlink and packages the target file contents into the Bento. The leaked file can then propagate further through export, push, or containerization workflows. An attacker can exfiltrate local files from the build host into the Bento artifact, exposing secrets such as cloud credentials, SSH keys, API tokens, environment files, or other sensitive local configurations. Because Bento artifacts are commonly exported, uploaded, stored, or containerized after build, the leaked file contents can spread beyond the original build machine. This issue has been fixed in version 1.4.39. | May 22, 2026 |
| CVE-2026-40607(opens NVD record) | Unscored | — | Mantis Bug Tracker (MantisBT) is an open source issue tracker. In versions 2.11.0 through 2.28.1, a Stored XSS vulnerability is caused by incorrect escaping of a saved filter's owner, allowing an attacker to inject arbitrary HTML on systems where $g_show_user_realname = ON. Note that By default, only users with Manager access level or above can save their filters publicly. This issue has been fixed in version 2.28.2. If developers are unable to update immediately, they can work around this issue by preventing display of users' real names (set $g_ show_user_realname = OFF; in configuration), and restricting the ability to store filters (set $g_stored_query_create_threshold / $g_stored_query_create_shared_threshold to NOBODY). | May 22, 2026 |