Vulnerability database
Every flaw the U.S. government’s cybersecurity agency (CISA) has confirmed attackers are using, plus the new WordPress plugin and theme bugs published each day, sorted by what they mean for a small business. We refresh it twice a day. If you see something your website or office runs, tell us.
1,839 results · page 3 of 37
Oct 5, 2026vLLM is an inference and serving engine for large language models.CVE-2026-105756 · Other software · vllm-project · reported by GitHubMedium 6.5
vLLM is an inference and serving engine for large language models. Prior to 0.30.0, OpenAI-compatible request models accept a non-empty cache_salt value without enforcing the character and length restrictions required by the IPCCacheServerKey consumer in LMCache-MP. On deployments using the LMCache-MP connector, a salt that contains a forbidden character or exceeds the permitted length can raise an uncaught ValueError during scheduler cache lookup, causing EngineCore to terminate and denying service to all concurrent users. This issue is fixed in version 0.30.0.
- Product
- vllm
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 5, 2026vLLM is an inference and serving engine for large language models.CVE-2026-105755 · Other software · vllm-project · reported by GitHubMedium 4.2
vLLM is an inference and serving engine for large language models. Prior to 0.30.0, flash late-interaction scoring at the /score and /rerank endpoints derives each worker's query_key value from the caller-controlled X-Request-Id header. A concurrent request that reuses a victim's identifier can overwrite the cached query embedding so the victim's documents are scored against the attacker's query, and shared use counters can also cause a late-interaction cache-miss error. This issue is fixed in version 0.30.0.
- Product
- vllm
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 5, 2026vLLM is an inference and serving engine for large language models.CVE-2026-105754 · Other software · vllm-project · reported by GitHubMedium 6.5
vLLM is an inference and serving engine for large language models. Prior to 0.30.0, the /inference/v1/generate endpoint in the disaggregated scale-out path accepts caller-supplied tensors in the features.kwargs_data field, cache identifiers in the features.mm_hashes field, ranges in the features.mm_placeholders field, and wire-selected multimodal field processors without rebinding them to the active model renderer contract. Forged grid geometry, field types, or non-positive placeholder lengths can terminate the shared EngineCore; when an attacker knows or can induce a victim's content hash, forged cache hashes can poison or retrieve cross-request encoder-cache state; and dropped sparse placeholder masks can alter replayed transport semantics. This issue is fixed in version 0.30.0.
- Product
- vllm
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 5, 2026vLLM is an inference and serving engine for large language models.CVE-2026-105753 · Other software · vllm-project · reported by GitHubMedium 6.5
vLLM is an inference and serving engine for large language models. Prior to 0.28.0, the default mirrored multimodal LRU cache can commit a media hash in the frontend sender cache during multimodal rendering and before engine admission, while the engine receiver cache never receives the payload if that request is rejected. A later request reusing the same media hash causes MultiModalProcessorSenderCache to send no payload and MultiModalReceiverCache to reach an assertion with the message "Expected a cached item," producing a shared-service availability failure. This issue is fixed in version 0.28.0.
- Product
- vllm
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 5, 2026vLLM is an inference and serving engine for large language models.CVE-2026-105752 · Other software · vllm-project · reported by GitHubLow 3.1
vLLM is an inference and serving engine for large language models. Prior to 0.30.0, Harmony tool continuations submitted through "POST /v1/responses" requests rebuild the next-turn engine input without preserving the cache_salt value, placing the continuation prefix in the global unsalted cache namespace even when the caller enabled salting. On deployments with prefix caching enabled, which is the default, an authenticated tenant who can reconstruct a victim's low-entropy post-tool history can submit the same continuation and use the cached_tokens_per_turn count to determine whether the prefix was previously processed, defeating the intended tenant isolation of salted prefix caching. This issue is fixed in version 0.30.0.
- Product
- vllm
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 2, 2026crmne/ruby_llm at commit fa6f279847d6d7027814539d9c0dfc3bbdfd2a83 contains a polynomial-time regular expression denial-of-service condition in Mistral model capCVE-2026-67989 · Other software · reported by MITREHigh 7.5
crmne/ruby_llm at commit fa6f279847d6d7027814539d9c0dfc3bbdfd2a83 contains a polynomial-time regular expression denial-of-service condition in Mistral model capability matching on Ruby 3.1.x
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 2, 2026MISP contains an improper input validation vulnerability in the decaying model import functionality.CVE-2026-104908 · Other software · MISP · reported by 5a6e4751-2f3f-4070-9419-94fb35b644e8High 7.1
MISP contains an improper input validation vulnerability in the decaying model import functionality. The import endpoint was intended to create a new decaying model belonging exclusively to the importing user's organisation, with the default flag forced to off. However, the application stripped only the top-level id and uuid fields and pinned org_id and default on the outer array before saving the data flat. A user with decaying-model permissions could supply a nested model key carrying its own primary key, organisation identifier, and default flag, which bypassed those guards during the save operation. Impact: - A user with perm_decaying could overwrite an existing decaying model belonging to another organisation in place, altering its name, formula, parameters, or ownership. - A user could create or modify a model flagged as the organisation default, affecting scoring behaviour for other users. - A user could reassign a model's organisation to an arbitrary value. Preconditions: - Authenticated user with decaying-model permission (perm_decaying). - Network access to the MISP instance. Affected: <2.5.48.
- Product
- MISP
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 1, 2026A security vulnerability has been detected in modelcontextprotocol mcp-server-fetch and mcp-server-everything up to 2026.6.4.CVE-2026-104120 · Other software · modelcontextprotocol · reported by VulDBMedium 5.5
A security vulnerability has been detected in modelcontextprotocol mcp-server-fetch and mcp-server-everything up to 2026.6.4. Affected is the function fetch_url of the file mcp_server_fetch/server.py of the component Fetch Tool. The manipulation of the argument url/path leads to server-side request forgery. The attack may be initiated remotely. The exploit has been disclosed publicly and may be used. The pull request to fix this issue awaits acceptance.
- Product
- mcp-server-fetch
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 1, 2026A path traversal vulnerability exists in the web management interface of multiple Multifunction Devices and Printers, including Apeos C4571 1.1.3 and earlier, ACVE-2026-78249 · Other software · Fujifilm Business Innovation Corp. · reported by 47e4fddc-a2f1-48a0-beab-943c84c1c954Medium 6.8
A path traversal vulnerability exists in the web management interface of multiple Multifunction Devices and Printers, including Apeos C4571 1.1.3 and earlier, Apeos C3567 1.1.3, or other products listed, specifically in the handling of externally supplied parameters. If the device receives a specially crafted, malicious request, it may trigger unintended processing.
- Product
- Apeos 3060 / 2560 / 1860 Japan model
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordglobal.sharpfujifilm.com
Oct 1, 2026Comelit Multi-User Gateway for VIP System (model 1456B) firmware versions 2.9.1 and 2.10.0 expose a network-accessible management interface that does not requirCVE-2026-80276 · Other software · Comelit Group S.p.A. · reported by db4dfee8-a97e-4877-bfae-eba6d14a2166High 7.5
Comelit Multi-User Gateway for VIP System (model 1456B) firmware versions 2.9.1 and 2.10.0 expose a network-accessible management interface that does not require authentication. Through this interface, sensitive device configuration data - including the Remote Configuration Password - can be read in cleartext by a remote, unauthenticated attacker.
- Product
- 1456B Multi-User Gateway
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordaccesscontrol.failpro.comelitgroup.com
Oct 1, 2026Comelit Multi-User Gateway for VIP System (model 1456B) firmware versions 2.9.1 and 2.10.0 fail to enforce server-side authorization on an administrative passwoCVE-2026-80275 · Other software · Comelit Group S.p.A. · reported by db4dfee8-a97e-4877-bfae-eba6d14a2166High 8.8
Comelit Multi-User Gateway for VIP System (model 1456B) firmware versions 2.9.1 and 2.10.0 fail to enforce server-side authorization on an administrative password-change function. An authenticated user level can invoke this function to overwrite the installer (administrator) account password.
- Product
- 1456B Multi-User Gateway
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordaccesscontrol.failpro.comelitgroup.com
Sep 30, 2026An Active Debug Code vulnerability in certain ASUS router models allows a remote authenticated user, via a crafted HTTP request, to bypass security mechanisms aCVE-2026-13313 · Network & remote access · ASUS · reported by 54bf65a7-a193-42d2-b1ba-8e150d3c35e1High 8.9
An Active Debug Code vulnerability in certain ASUS router models allows a remote authenticated user, via a crafted HTTP request, to bypass security mechanisms and enable the Telnet service, thereby executing arbitrary commands with root privileges and potentially affecting other devices connected to the router. Refer to the ' Security Update for ASUS Router Firmware ' section on the ASUS Security Advisory for more information.
- Product
- Router
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 30, 2026In camel-ai camel 0.2.91a1, v0.2.91a2 and v0.2.91a3, CodeExecutionToolkit can run model-produced Python code through SubprocessInterpreter without an approval bCVE-2026-51857 · Other software · reported by MITRECritical 9.8
In camel-ai camel 0.2.91a1, v0.2.91a2 and v0.2.91a3, CodeExecutionToolkit can run model-produced Python code through SubprocessInterpreter without an approval boundary.
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordgist.github.comgithub.com
Sep 30, 2026modelscope Agentscope v1.0.0-v1.0.8 is vulnerable to Path Traversal in insert_text_file.CVE-2026-51570 · Other software · reported by MITREHigh 8.1
modelscope Agentscope v1.0.0-v1.0.8 is vulnerable to Path Traversal in insert_text_file.
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 30, 2026modelscope Agentscope v1.0.18-v1.0.0 is vulnerable to Path Traversal in write_text_file.CVE-2026-51568 · Other software · reported by MITREHigh 8.1
modelscope Agentscope v1.0.18-v1.0.0 is vulnerable to Path Traversal in write_text_file.
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordgist.github.comgithub.com
Sep 30, 2026Inspur Haiyue HCM Cloud contains an arbitrary file read vulnerability in the /api/model_report/file/download endpoint that allows unauthenticated remote attackeCVE-2024-58387 · Other software · Inspur · reported by VulnCheckHigh 8.7
Inspur Haiyue HCM Cloud contains an arbitrary file read vulnerability in the /api/model_report/file/download endpoint that allows unauthenticated remote attackers to read arbitrary files by supplying unvalidated path parameters index and ext. Attackers can craft requests such as /api/model_report/file/download?index=/&ext=<path> to traverse the filesystem and disclose sensitive files including /etc/passwd, application database files, and system configuration files. Exploitation evidence was first observed by the Shadowserver Foundation on 2024-11-04 .
- Product
- Haiyue HCM Cloud
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 30, 2026LightLLM through 1.2.0 visual_only deployments expose an unauthenticated RPyC service with allow_pickle enabled that deserializes attacker-supplied arguments inCVE-2026-103395 · Other software · ModelTC · reported by VulnCheckCritical 9.3
LightLLM through 1.2.0 visual_only deployments expose an unauthenticated RPyC service with allow_pickle enabled that deserializes attacker-supplied arguments in the remote_infer_images method. Attackers can reach the visual RPyC port and pass objects with __reduce__ methods to execute arbitrary code with service account privileges.
- Product
- LightLLM
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 30, 2026LightLLM through 1.2.0 mounts reinforcement learning control routes on the public HTTP API without authentication checks.CVE-2026-103270 · Other software · ModelTC · reported by VulnCheckHigh 8.7
LightLLM through 1.2.0 mounts reinforcement learning control routes on the public HTTP API without authentication checks. Unauthenticated attackers can call endpoints like /pause_generation, /abort_request, /flush_cache, and /init_weights_update_group to disrupt inference operations and wedge workers on deployments started with --enable_rl.
- Product
- LightLLM
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 30, 2026LightLLM through 1.2.0 fails to validate image_url and audio_url parameters in multimodal endpoints, allowing unauthenticated attackers to perform server-side rCVE-2026-103243 · Other software · ModelTC · reported by VulnCheckMedium 6.9
LightLLM through 1.2.0 fails to validate image_url and audio_url parameters in multimodal endpoints, allowing unauthenticated attackers to perform server-side request forgery. Attackers can supply arbitrary URLs to fetch internal resources, with vision model processing disclosing content or error responses revealing internal network topology.
- Product
- LightLLM
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 29, 2026LightLLM through 1.2.0 contains a memory exhaustion vulnerability in the NCCL control channel when started with --pd_trans_mode nccl, allowing unauthenticated aCVE-2026-103042 · Other software · ModelTC · reported by VulnCheckHigh 8.7
LightLLM through 1.2.0 contains a memory exhaustion vulnerability in the NCCL control channel when started with --pd_trans_mode nccl, allowing unauthenticated attackers to exhaust KV-transfer worker memory. Attackers can call the exposed_set_value method to store unbounded key-value pairs without size limits, causing the worker process to crash and triggering node failure.
- Product
- LightLLM
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 29, 2026LightLLM through 1.2.0 multimodal deployments expose an unauthenticated RPyC cache service with pickle deserialization enabled on all interfaces.CVE-2026-103041 · Other software · ModelTC · reported by VulnCheckCritical 9.3
LightLLM through 1.2.0 multimodal deployments expose an unauthenticated RPyC cache service with pickle deserialization enabled on all interfaces. Attackers can send crafted serialized objects to exposed cache methods to execute arbitrary code with service privileges.
- Product
- LightLLM
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 29, 2026LightLLM through 1.2.0 contains a remote code execution vulnerability in the router profiler service when started with --enable_profiling flag.CVE-2026-103040 · Other software · ModelTC · reported by VulnCheckCritical 9.3
LightLLM through 1.2.0 contains a remote code execution vulnerability in the router profiler service when started with --enable_profiling flag. The service exposes an unauthenticated RPyC server with pickle deserialization enabled, allowing attackers to execute arbitrary code by sending crafted serialized objects to the profiler command queue.
- Product
- LightLLM
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 29, 2026In Baicells Nova 430H, an unauthenticated device within radio range can send a malformed uplink message during connection setup that contains an invalid NAS payCVE-2026-96274 · Other software · Baicells · reported by ics-cert@hq.dhs.govHigh 8.3
In Baicells Nova 430H, an unauthenticated device within radio range can send a malformed uplink message during connection setup that contains an invalid NAS payload. Because the eNodeB does not properly validate this payload, it forwards the message to the core network, which can trigger a shutdown of the signaling association for the cell. This results in a temporary service disruption until the eNodeB and core network re-establish connectivity.
- Product
- Nova 430H eNodeB (model pBS3101SH)
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 29, 2026Attacker model / Preconditions: a loaded `TXM_MODULE_USER_MODE | TXM_MODULE_MEMORY_PROTECTION` module issuing kernel dispatch calls, on a build with `TX_ENABLE_CVE-2026-102710 · Other software · Eclipse Foundation · reported by emo@eclipse.orgCritical 9.3
Attacker model / Preconditions: a loaded `TXM_MODULE_USER_MODE | TXM_MODULE_MEMORY_PROTECTION` module issuing kernel dispatch calls, on a build with `TX_ENABLE_EVENT_TRACE`. A user-mode, memory-protected module can register an arbitrary function pointer as the global trace-full callback. The kernel calls it directly — no validation, no trampoline — from privileged kernel code when the trace buffer wraps. An invalid pointer faults the kernel (DoS). A pointer into the module's own code was observed running with kernel privilege (`CONTROL.nPRIV = 0`), confirmed at runtime with a register capture inside that code.
- Product
- eclipse-threadx/threadx
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 29, 2026Deserialization of untrusted data in the model loading component in Amazon GluonTS before 0.17.0 might allow context-dependent attackers to execute arbitrary opCVE-2026-100308 · Other software · AWS · reported by ff89ba41-3aa1-4d27-914a-91399e9639e5High 8.4
Deserialization of untrusted data in the model loading component in Amazon GluonTS before 0.17.0 might allow context-dependent attackers to execute arbitrary operating system commands with the privileges of the loading process via a crafted serialized model directory. To remediate this issue, users should upgrade to version 0.17.0 or later.
- Product
- gluonts
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordaws.amazon.comgithub.com
Sep 29, 2026CTranslate2 before 4.8.1 contains an out-of-bounds heap read vulnerability in the binary model loader when deserializing string fields without null terminators.CVE-2026-102567 · Other software · OpenNMT · reported by VulnCheckMedium 6.9
CTranslate2 before 4.8.1 contains an out-of-bounds heap read vulnerability in the binary model loader when deserializing string fields without null terminators. Attackers can craft malicious model files to trigger heap memory reads past buffer boundaries, causing crashes or disclosing adjacent heap memory contents.
- Product
- CTranslate2
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 29, 2026CTranslate2 before 4.8.1 contains a heap-based buffer overflow in the binary model loader that fails to validate payload length against allocated buffer size.CVE-2026-102566 · Other software · OpenNMT · reported by VulnCheckHigh 8.5
CTranslate2 before 4.8.1 contains a heap-based buffer overflow in the binary model loader that fails to validate payload length against allocated buffer size. Attackers can craft malicious model files with oversized payload lengths to write past heap allocation boundaries, causing crashes or arbitrary code execution.
- Product
- CTranslate2
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 29, 2026The Apache XmlSchema walker (xmlschema-walker) doesn't detect cycles in type derivation, substitution groups, model groups or attribute groups.CVE-2026-102497 · Websites & servers · Apache Software Foundation · reported by security@apache.orgHigh 7.5
The Apache XmlSchema walker (xmlschema-walker) doesn't detect cycles in type derivation, substitution groups, model groups or attribute groups. A malicious schema with such a cycle can make the walker recurse until the stack overflows, causing a denial of service. Users are recommended to upgrade to version 2.3.3, which fixes this issue.
- Product
- Apache XMLSchema
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordlists.apache.orgopenwall.com
Sep 29, 2026Apache XmlSchema doesn't limit how deeply schema structures can be nested when it builds its schema model, so a malicious schema can make parsing recurse until CVE-2026-102496 · Websites & servers · Apache Software Foundation · reported by security@apache.orgHigh 7.5
Apache XmlSchema doesn't limit how deeply schema structures can be nested when it builds its schema model, so a malicious schema can make parsing recurse until the stack overflows. This causes a denial of service. Users are recommended to upgrade to version 2.3.3, which fixes this issue.
- Product
- Apache XMLSchema
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordlists.apache.orgopenwall.com
Sep 28, 2026Unsloth Zoo versions 2025.9.9 before 2026.8.14, as implemented in Unsloth 2025.9.9 through 2026.8.19, contains a code injection vulnerability in the model-loadiCVE-2026-93348 · Other software · unslothai · reported by VulnCheckHigh 8.6
Unsloth Zoo versions 2025.9.9 before 2026.8.14, as implemented in Unsloth 2025.9.9 through 2026.8.19, contains a code injection vulnerability in the model-loading compile path where the get_transformers_model_type() function in hf_utils.py collects model_type values from nested model configurations without enforcing a character allowlist, allowing newlines and arbitrary Python source to survive normalization. Attackers can embed a newline in a nested model_type value within a malicious model's config.json to terminate the generated import statement and execute arbitrary Python code via exec() in unsloth_compile_transformers(), achieving remote code execution as the loading user when the model is loaded for training or inference.
- Product
- unsloth-zoo
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 26, 2026Hugo versions from v0.161.0 through v0.165.0 run Node.js tools (css.PostCSS, css.TailwindCSS, js.Babel) under the Node.js permission model to restrict file systCVE-2026-100690 · Other software · gohugoio · reported by VulnCheckHigh 8.7
Hugo versions from v0.161.0 through v0.165.0 run Node.js tools (css.PostCSS, css.TailwindCSS, js.Babel) under the Node.js permission model to restrict file system reads to the project directory and configured mounts. Because the Node.js permission model validates only the lexical path and follows symbolic links that point outside the allowed set, Hugo did not detect symlinks escaping the sandbox. An attacker who can contribute content to a Hugo project (for example via a pull request) can commit a symlink such as assets/css/x.css -> /etc/passwd together with a PostCSS plugin that reads it, allowing any file readable by the Hugo build process to be disclosed and potentially embedded in the published site. This affects builds using the default security configuration; projects that do not invoke Node.js tools are unaffected. Fixed in v0.166.0, which scans allowed paths and fails the build when a symbolic link resolves outside them.
- Product
- hugo
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordgithub.comvulncheck.com
Sep 26, 2026vLLM is an inference and serving engine for large language models.CVE-2026-100653 · Other software · vllm-project · reported by VulnCheckHigh 8.3
vLLM is an inference and serving engine for large language models. In versions from 0.22.1 through 0.28.0, the operator-supplied model revision pin (--revision / --code-revision) is not propagated to several Hugging Face artifact loads for the FunAudioChat and Tarsier2 architectures: the WhisperFeatureExtractor and speech_tokenizer PreTrainedTokenizerFast loads in vllm/model_executor/models/funaudiochat.py and the Qwen2VLConfig.from_pretrained call used by Tarsier2ProcessingInfo in vllm/model_executor/models/qwen2_vl.py. As a result, deployments pinned to a reviewed revision still resolve these behavior-affecting processor, tokenizer, and config artifacts from the repository's default revision, so a later change to the upstream default branch can alter audio preprocessing, speech tokenizer behavior, or Tarsier2 configuration without any change to the operator's configured pin. This is a supply-chain integrity and reproducibility failure for pinned deployments; it is residual to the earlier fix tracked as GHSA-3ww4-5jv9-j5gm / CVE-2026-47155 and does not constitute remote code execution or a trust_remote_code=False bypass. The issue is fixed in version 0.28.0.
- Product
- vllm
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 25, 2026OpenClaw (npm package 'openclaw') before 2026.7.1 improperly handles case sensitivity in the model-facing cron tool: a mixed-case payload kind can pass the agenCVE-2026-100580 · Other software · OpenClaw · reported by VulnCheckHigh 8.7
OpenClaw (npm package 'openclaw') before 2026.7.1 improperly handles case sensitivity in the model-facing cron tool: a mixed-case payload kind can pass the agent-facing shell-execution guard and later normalize into a command job. An actor able to steer a tool-enabled agent can therefore create a persistent cron job that executes attacker-selected commands with the privileges of the OpenClaw process user, resulting in access to host files and credentials and impact to scheduled service availability. The issue is limited to cron jobs created or edited through the model-facing cron tool; direct CLI and authorized Gateway scheduling surfaces are trusted operator controls. Fixed in 2026.7.1.
- Product
- OpenClaw
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordgithub.comvulncheck.com
Sep 25, 2026OpenClaw versions before 2026.8.1 fail to properly restrict access to operator command cron jobs, allowing model-visible agent callers to read and execute ownerCVE-2026-100568 · Other software · OpenClaw · reported by VulnCheckHigh 8.7
OpenClaw versions before 2026.8.1 fail to properly restrict access to operator command cron jobs, allowing model-visible agent callers to read and execute ownerless command jobs. Attackers can inspect stored environment variables and force-run disabled or unscheduled command jobs to access secrets and execute operator-authored commands.
- Product
- OpenClaw
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordgithub.comvulncheck.com
Sep 25, 2026OpenClaw Feishu before 2026.8.1 fails to validate whether a configured default account is disabled before selecting it for model tool operations.CVE-2026-100540 · Other software · openclaw · reported by VulnCheckHigh 7.6
OpenClaw Feishu before 2026.8.1 fails to validate whether a configured default account is disabled before selecting it for model tool operations. Attackers can exploit multi-account setups where a disabled default account retains credentials to read or modify Feishu resources through a revoked identity.
- Product
- feishu
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordgithub.comvulncheck.com
Sep 25, 2026In the Linux kernel, the following vulnerability has been resolved: bpf: Reject legacy packet loads from callbacks check_ld_abs() models a failed BPF_LD_ABS oCVE-2026-98044 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67
In the Linux kernel, the following vulnerability has been resolved: bpf: Reject legacy packet loads from callbacks check_ld_abs() models a failed BPF_LD_ABS or BPF_LD_IND in a subprogram as an implicit return with R0 set to zero. It calls prepare_func_exit() to explore this synthesized path. When the load is reached directly from a synchronous callback, prepare_func_exit() enforces the callback return contract and marks R0 precise. R0 is not derived from a real instruction on this path, so precision backtracking reaches the callback call with R0 still requested and triggers the "callback unexpected regs" verifier bug. A privileged program loader can therefore cause a verifier warning and an -EFAULT BPF_PROG_LOAD. These legacy packet-load instructions are deprecated. Reject them from callbacks rather than complicating their implicit-return model. Check all active frames before constructing the implicit return so nested static subprograms cannot hide the callback context. Global functions are verified independently with a fresh frame zero, so an active-frame check cannot identify a global function called from a callback. Also check the complete subprogram call graph during stack
- Product
- Linux
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordgit.kernel.orggit.kernel.org
Sep 23, 2026JobJobEventsChildrenSummary view has no model/parent_model.CVE-2026-71459 · Other software · Red Hat · reported by Red HatMedium 5
JobJobEventsChildrenSummary view has no model/parent_model. ModelAccessPermission.check_get_permissions() falls through (returns True) for any authenticated user. The view uses raw get_object_or_404(Job, pk) without DRF object-level permission check. Zero-privilege user reads event tree structure, event_processing_finished status, and enumerates Job IDs platform-wide via 200/404 oracle. Sibling endpoint /jobs/{id}/job_events/ correctly returns 403.
- Product
- Red Hat Ansible Automation Platform 2.5 for RHEL 8
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordaccess.redhat.comaccess.redhat.com
Sep 23, 2026URLModificationMiddleware resolves named-URL lookups against unfiltered Model.objects before RBAC.CVE-2026-71458 · Other software · Red Hat · reported by Red HatMedium 5
URLModificationMiddleware resolves named-URL lookups against unfiltered Model.objects before RBAC. The 403→404 shim only rewrites 403 responses, leaving the pk=0 miss path with a different 404 detail string. Differential "Not found." vs "No <Model> matches..." reveals whether a named resource (org, credential, inventory, host) exists anywhere on the platform. Enables cross-tenant internal hostname enumeration.
- Product
- Red Hat Ansible Automation Platform 2.5 for RHEL 8
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordaccess.redhat.comaccess.redhat.com
Sep 23, 2026MLflow's statsmodel flavor, versions 2.1.0 to 3.14.0, omits the MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False security control entirely in _load_model(), which alloCVE-2026-96804 · Other software · MLflow · reported by cret@cert.orgHigh 8.8
MLflow's statsmodel flavor, versions 2.1.0 to 3.14.0, omits the MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False security control entirely in _load_model(), which allows a remote attacker to execute arbitrary code via a crafted MLmodel artifact.
- Product
- MLflow
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 23, 2026MLflow's dspy flavor, versions >= 2.0, applies the MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False security control only when the model_path ends in .pkl, which alloCVE-2026-96775 · Other software · MLflow · reported by cret@cert.orgHigh 8.8
MLflow's dspy flavor, versions >= 2.0, applies the MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False security control only when the model_path ends in .pkl, which allows a remote attacker to execute arbitrary code via a crafted MLmodel artifact.
- Product
- MLflow
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 23, 2026LightLLM through 1.2.0 contains a remote code execution vulnerability in the KV-transfer worker when started with --pd_trans_mode nccl, which exposes an unautheCVE-2026-96560 · Other software · ModelTC · reported by VulnCheckCritical 9.3
LightLLM through 1.2.0 contains a remote code execution vulnerability in the KV-transfer worker when started with --pd_trans_mode nccl, which exposes an unauthenticated RPyC control channel that deserializes attacker-supplied data. Attackers can send malicious pickled objects to the exposed RPyC ThreadedServer to execute arbitrary code with the privileges of the LightLLM service account.
- Product
- LightLLM
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 22, 2026MCMS 6.1.1 through 6.2.1 has a SQL injection vulnerability in the custom model/form import feature.CVE-2026-88416 · Other software · reported by MITRECritical 9.8
MCMS 6.1.1 through 6.2.1 has a SQL injection vulnerability in the custom model/form import feature.
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 22, 2026Substance3D - Modeler is affected by an out-of-bounds write vulnerability that could result in arbitrary code execution in the context of the current user.CVE-2026-83963 · Other software · Adobe · reported by AdobeHigh 7.8
Substance3D - Modeler is affected by an out-of-bounds write vulnerability that could result in arbitrary code execution in the context of the current user. Exploitation of this issue requires user interaction in that a victim must open a malicious file.
- Product
- Substance3D - Modeler
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 22, 2026Substance3D - Modeler is affected by a Stack-based Buffer Overflow vulnerability that could result in arbitrary code execution in the context of the current useCVE-2026-83962 · Other software · Adobe · reported by AdobeHigh 7.8
Substance3D - Modeler is affected by a Stack-based Buffer Overflow vulnerability that could result in arbitrary code execution in the context of the current user. Exploitation of this issue requires user interaction in that a victim must open a malicious file.
- Product
- Substance3D - Modeler
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 22, 2026Substance3D - Modeler is affected by an out-of-bounds write vulnerability that could result in arbitrary code execution in the context of the current user.CVE-2026-81998 · Other software · Adobe · reported by AdobeHigh 7.8
Substance3D - Modeler is affected by an out-of-bounds write vulnerability that could result in arbitrary code execution in the context of the current user. Exploitation of this issue requires user interaction in that a victim must open a malicious file.
- Product
- Substance3D - Modeler
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 22, 2026Substance3D - Modeler is affected by an out-of-bounds write vulnerability that could result in arbitrary code execution in the context of the current user.CVE-2026-79906 · Other software · Adobe · reported by AdobeHigh 7.8
Substance3D - Modeler is affected by an out-of-bounds write vulnerability that could result in arbitrary code execution in the context of the current user. Exploitation of this issue requires user interaction in that a victim must open a malicious file.
- Product
- Substance3D - Modeler
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 22, 2026MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira).CVE-2026-77272 · Other software · sooperset · reported by GitHubMedium 5.4
MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira). Prior to 0.22.0, the OAuth error query parameter is passed to CallbackHandler._send_response in oauth_setup.py and interpolated into an HTML page without escaping. A crafted authorization callback can inject markup or script that executes in the browser of a user completing the OAuth flow. This issue is fixed in version 0.22.0.
- Product
- mcp-atlassian
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 22, 2026MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira).CVE-2026-77269 · Other software · sooperset · reported by GitHubMedium 6.5
MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira). Prior to 0.22.0, the remediation for CVE-2026-27825 protects download destinations but does not constrain source paths used by attachment uploads. A caller can provide an absolute or traversal file_path and cause the server to upload the selected local file. The advisory traces the vulnerable input and processing flow through upload_attachment, file_path, and CVE-2026-27825, which identify the affected entry points, controls, and code paths. This issue is fixed in version 0.22.0.
- Product
- mcp-atlassian
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 22, 2026MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira).CVE-2026-77268 · Other software · sooperset · reported by GitHubMedium 5.5
MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira). Prior to 0.22.0, the OAuth fallback token directory and JSON file are created without explicit owner-only modes. Local users or processes with access through the resulting group or world permission bits can read access and refresh tokens and reuse the associated Atlassian session. The advisory traces the vulnerable input and processing flow through ~/.mcp-atlassian, oauth-<client_id>.json, access_token, and refresh_token, which identify the affected entry points, controls, and code paths. This issue is fixed in version 0.22.0.
- Product
- mcp-atlassian
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 22, 2026MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira).CVE-2026-77266 · Other software · sooperset · reported by GitHubMedium 6.5
MCP Atlassian is a Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira). Prior to 0.22.0, upload_attachment accepts absolute paths and traversal sequences without constraining the resolved path to the server workspace. An MCP caller with attachment access can read a chosen server-local file and exfiltrate it through Jira or Confluence. The advisory traces the vulnerable input and processing flow through upload_attachment, file_path, and path traversal, which identify the affected entry points, controls, and code paths. This issue is fixed in version 0.22.0.
- Product
- mcp-atlassian
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Where this comes from: the exploited list is CISA’s Known Exploited Vulnerabilities catalog. Everything else is the whole National Vulnerability Database, where each record is published by the team that found the bug (Wordfence, Patchstack, WPScan, the vendors themselves and others); we show who. Exploit odds are FIRST’s EPSS, updated daily. Titles and categories are ours, so check the linked record before acting on a detail. For a write-up of the big ones, see our advisories. Every vendor, and every WordPress plugin and theme with a recorded flaw, also has its own page. All of it is free to reuse as JSON, CSV or RSS.
Weekly security brief
The flaws that hit local websites, every Friday.
One short email: what got exploited this week, which WordPress plugins broke, and what to check. Unsubscribe in one click.
Tell us what you're building.
A website, your marketing, a security check, an app, or the thing you've been running out of a spreadsheet for three years. A few lines is plenty.
Not ready to talk yet? Run the free website check and see what Google sees.