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.
715 results · page 3 of 15
Aug 23, 2026A vulnerability was determined in ggml-org llama.cpp bec4772f6.CVE-2026-78148 · Other software · ggml-org · reported by VulDBMedium 6.9
A vulnerability was determined in ggml-org llama.cpp bec4772f6. This affects the function rpc_server::graph_compute of the file ggml/src/ggml-rpc/ggml-rpc.cpp of the component ggml-RPC Server. Executing a manipulation can lead to null pointer dereference. The attack may be launched remotely. The pull request to fix this issue awaits acceptance.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 23, 2026A vulnerability was found in ggml-org llama.cpp bec4772f6.CVE-2026-78147 · Other software · ggml-org · reported by VulDBMedium 6.9
A vulnerability was found in ggml-org llama.cpp bec4772f6. The impacted element is the function deserialize_tensor of the file ggml/src/ggml-rpc/ggml-rpc.cpp of the component ggml-RPC Server. Performing a manipulation of the argument op/op_params results in deserialization. The attack may be initiated remotely. This vulnerability is distinct from CVE-2026-34159 (GHSA-j8rj-fmpv-wcxw, PR #20908), which only added a buffer==nullptr rejection in create_node() and does not validate op or op_params. The reported GitHub issue was closed automatically due to inactivity.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 22, 2026In the Linux kernel, the following vulnerability has been resolved: hwmon: (ltc4282) Clamp negative current limits When a negative value is passed to ltc4282_CVE-2026-74685 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67
In the Linux kernel, the following vulnerability has been resolved: hwmon: (ltc4282) Clamp negative current limits When a negative value is passed to ltc4282_write_curr(), the signed long val is cast directly to u64: drivers/hwmon/ltc4282.c:ltc4282_write_curr() { /* need to pass it in millivolt */ u32 in = DIV_ROUND_CLOSEST_ULL((u64)val * st->rsense, DECA * MICRO); ... } This cast converts negative inputs into large positive values. The subsequent division result overflows the u32 in variable, truncating to a pseudo-random positive value. When this is passed to ltc4282_write_voltage_byte(), it is clamped to the maximum limit instead of zero. Clamp val to 0 and to the maximum supported upper limit before the cast and assign the result to a 64-bit temporary variable before the division to avoid the underflow and an also possible overflow.
- 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
Aug 21, 2026In the Neptune connector, a user with access to Neptune through Athena Federated Query could gain access to properties in the Lambda supplying the compute for tCVE-2026-77810 · Other software · AWS · reported by ff89ba41-3aa1-4d27-914a-91399e9639e5Critical 9.4
In the Neptune connector, a user with access to Neptune through Athena Federated Query could gain access to properties in the Lambda supplying the compute for the connector. To remediate this issue, users should upgrade to aws-athena-query-federation v2026.30.1 or later.
- Product
- Athena Federated Query Neptune Connector
- 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
Aug 20, 2026Dockge validates a stack name only on the write path.CVE-2026-73040 · Other software · louislam · reported by VulnCheckHigh 8.7
Dockge validates a stack name only on the write path. In backend/stack.ts the allow-list check in validate(), which requires the name to match ^[a-z0-9_-]+$, is reached from save() alone, while the path getter returns path.join(this.server.stacksDir, this.name) and Stack.getStack builds path.join(server.stacksDir, stackName) with no check. The socket handlers in backend/agent-socket-handlers/docker-socket-handler.ts confirm the caller is logged in and that the name is a string, then pass it straight to Stack.getStack, so a name containing traversal sequences resolves outside the managed stacks directory. An authenticated user can therefore read the composeENV and composeYAML values of any directory the server process can reach, which discloses the secrets in that directory's .env or Compose file, and can invoke delete(), which runs docker compose down and then fsAsync.rm on the traversed path with recursive and force set, removing that directory. Disclosure is limited to files named .env or an accepted Compose filename, and deletion requires the target directory to hold a valid Compose file so that docker compose down exits successfully. Dockge commonly runs as root with access to
- Product
- dockge
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 18, 2026CodeWhale versions before 0.8.64 contain an argument injection vulnerability in the git_blame tool that allows attackers to read arbitrary files by injecting giCVE-2026-75912 · Other software · Hmbown · reported by VulnCheckHigh 8.3
CodeWhale versions before 0.8.64 contain an argument injection vulnerability in the git_blame tool that allows attackers to read arbitrary files by injecting git options into the unvalidated rev parameter. Attackers can supply rev values like --contents=/path/to/file to exfiltrate sensitive files such as SSH keys and credentials through the tool output returned to the model.
- Product
- CodeWhale
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 15, 2026In the Linux kernel, the following vulnerability has been resolved: ALSA: usb-audio: Clamp frame size in implicit-feedback mode snd_usb_handle_sync_urb() scalCVE-2026-74497 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67High 8.4
In the Linux kernel, the following vulnerability has been resolved: ALSA: usb-audio: Clamp frame size in implicit-feedback mode snd_usb_handle_sync_urb() scales received sync packet sizes by the sender's stride and stores the result directly in out_packet->packet_size[i]. If a connected USB device sends an oversized sync packet, this frame count can exceed ep->maxframesize. The un-clamped frame count then propagates to the playback endpoint queue, potentially driving packet transfers beyond the endpoint's hardware frame limits. Cap the calculated frame count against ep->maxframesize in snd_usb_handle_sync_urb() to prevent oversized packets from entering the playback queue.
- 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
Aug 15, 2026In the Linux kernel, the following vulnerability has been resolved: net: watchdog: fix refcount tracking races Blamed commit converted the untracked dev_hold(CVE-2026-74264 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67High 7.8
In the Linux kernel, the following vulnerability has been resolved: net: watchdog: fix refcount tracking races Blamed commit converted the untracked dev_hold()/dev_put() calls in the watchdog code to use the tracked dev_hold_track()/dev_put_track() (which were later renamed/interfaced to netdev_hold() and netdev_put()). By introducing dev->watchdog_dev_tracker to store the reference tracking information without adding synchronization between netdev_watchdog_up() and dev_watchdog(), it enabled the race condition where this pointer could be overwritten or freed concurrently, leading to the list corruption crash syzbot reported: list_del corruption, ffff888114a18c00->next is NULL kernel BUG at lib/list_debug.c:52 ! Oops: invalid opcode: 0000 [#1] SMP KASAN PTI CPU: 1 UID: 0 PID: 91 Comm: kworker/u8:5 Not tainted syzkaller #0 PREEMPT(lazy) Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/09/2026 Workqueue: events_unbound linkwatch_event RIP: 0010:__list_del_entry_valid_or_report.cold+0x22/0x2a lib/list_debug.c:52 Call Trace: <TASK> __list_del_entry_valid include/linux/list.h:132 [inline] __list_del_entry include/linux/list.h:246 [inline] li
- 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
Aug 15, 2026In the Linux kernel, the following vulnerability has been resolved: tcp: ipv6: clamp default adverting MSS to avoid GSO_BY_FRAGS (0xFFFF) When MTU is large, iCVE-2026-72502 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67High 7.5
In the Linux kernel, the following vulnerability has been resolved: tcp: ipv6: clamp default adverting MSS to avoid GSO_BY_FRAGS (0xFFFF) When MTU is large, ip6_default_advmss() can return IPV6_MAXPLEN (65535). This is interpreted by TCP as mss_clamp, allowing the MSS to reach 65535. However, 0xFFFF is also used as a magic value GSO_BY_FRAGS in the kernel. If a TCP packet with gso_size=0xFFFF is passed to skb_segment(), it will be mistakenly treated as GSO_BY_FRAGS, leading to a NULL pointer dereference because local TCP packets do not use frag_list. Fix this by returning min(IPV6_MAXPLEN, GSO_BY_FRAGS - 1) (65534) from ip6_default_advmss() when MTU is large. Also update the stale comment in ip6_default_advmss() which suggested that IPV6_MAXPLEN is returned to mean "any MSS".
- 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
Aug 13, 2026A vulnerability was determined in dromara lamp-cloud up to 5.10.0.CVE-2026-19758 · Other software · dromara · reported by VulDBMedium 5.5
A vulnerability was determined in dromara lamp-cloud up to 5.10.0. This issue affects some unknown processing of the file FileChunkController.java of the component chunk-check endpoint. Executing a manipulation of the argument Name can lead to path traversal. The attack can be launched remotely. The exploit has been publicly disclosed and may be utilized. The project was informed of the problem early through an issue report but has not responded yet.
- Product
- lamp-cloud
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 13, 2026A vulnerability was found in Dromara lamp-cloud up to 5.10.0.CVE-2026-19757 · Other software · Dromara · reported by VulDBMedium 5.5
A vulnerability was found in Dromara lamp-cloud up to 5.10.0. This vulnerability affects unknown code of the file FileAnyoneController.java of the component File-Upload Controller. Performing a manipulation of the argument bucket/bizType results in path traversal. The attack can be initiated remotely. The exploit has been made public and could be used. The project was informed of the problem early through an issue report but has not responded yet.
- Product
- lamp-cloud
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 13, 2026A vulnerability has been found in Dromara lamp-cloud up to 5.10.0.CVE-2026-19756 · Other software · Dromara · reported by VulDBLow 2.1
A vulnerability has been found in Dromara lamp-cloud up to 5.10.0. This affects an unknown part of the file DefGenProjectController.java of the component Code Generator. Such manipulation of the argument outputDir/parent/projectPrefix leads to path traversal. It is possible to launch the attack remotely. The exploit has been disclosed to the public and may be used. The project was informed of the problem early through an issue report but has not responded yet.
- Product
- lamp-cloud
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 11, 2026In the Linux kernel, the following vulnerability has been resolved: drm/amdkfd: clamp v9 CRIU control stack checkpoint copy to BO size CRIU checkpoint copies CVE-2026-68447 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67High 7.1
In the Linux kernel, the following vulnerability has been resolved: drm/amdkfd: clamp v9 CRIU control stack checkpoint copy to BO size CRIU checkpoint copies the MQD control stack using cp_hqd_cntl_stack_size from hardware without bounding it to the allocated BO region. If the HW field is larger than the queue's control stack allocation, memcpy reads past the BO into adjacent GTT memory and can leak kernel data to userspace. Store the page-aligned control stack BO size in mqd_manager and clamp checkpoint copies and reported checkpoint sizes to min(cp_hqd_cntl_stack_size, mm->ctl_stack_size). Apply the same bound for multi-XCC v9.4.3 checkpoint layout. (cherry picked from commit 6c2abd0ec09e86c6323010673766f76050e28aa3)
- 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
Aug 10, 2026In the Linux kernel, the following vulnerability has been resolved: wifi: mac80211_hwsim: clamp virtio RX length before skb_put hwsim_virtio_rx_work() passes CVE-2026-68411 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67
In the Linux kernel, the following vulnerability has been resolved: wifi: mac80211_hwsim: clamp virtio RX length before skb_put hwsim_virtio_rx_work() passes the virtqueue used-ring length reported by the device straight to skb_put() on a fixed-size receive skb. A backend reporting a length larger than the skb tailroom drives skb_put() past the buffer end and hits skb_over_panic() -- a host-triggerable guest panic (denial of service). Clamp the length to the skb's available room before skb_put(). A conforming device never reports more than the posted buffer size, so valid frames are unaffected; a truncated over-report then fails the length/header checks in hwsim_virtio_handle_cmd() and is dropped, so truncating rather than dropping here cannot be turned into a parsing problem.
- 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
Aug 10, 2026In the Linux kernel, the following vulnerability has been resolved: usb: gadget: uvc: clamp SEND_RESPONSE length to the response buffer uvc_send_response() buCVE-2026-68366 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67
In the Linux kernel, the following vulnerability has been resolved: usb: gadget: uvc: clamp SEND_RESPONSE length to the response buffer uvc_send_response() builds the UVC control response from a user-supplied struct uvc_request_data: req->length = min_t(unsigned int, uvc->event_length, data->length); ... memcpy(req->buf, data->data, req->length); req->length is clamped to uvc->event_length, which is taken from the host control request wLength (up to UVC_MAX_REQUEST_SIZE, 64), and to data->length, which comes from the UVCIOC_SEND_RESPONSE ioctl and is only checked for being negative. The source buffer data->data is only 60 bytes, so a response with uvc->event_length and data->length both greater than 60 makes memcpy() read past the end of data->data. Clamp req->length to sizeof(data->data) as well.
- 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
Aug 9, 2026A flaw has been found in NightTrek Ollama-mcp up to 80cf2e17cfc144963a475b619093a2d13c13dbc9.CVE-2026-19334 · Other software · NightTrek · reported by VulDBLow 1.9
A flaw has been found in NightTrek Ollama-mcp up to 80cf2e17cfc144963a475b619093a2d13c13dbc9. This affects an unknown part of the file src/index.ts. This manipulation of the argument name/modelfile/source/destination causes command injection. The attack can only be executed locally. This product is using a rolling release to provide continious delivery. Therefore, no version details for affected nor updated releases are available. The project was informed of the problem early through an issue report but has not responded yet.
- Product
- Ollama-mcp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 7, 2026A vulnerability in the XAR file format parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition or possibly other expanded impaCVE-2026-20348 · Network & remote access · Cisco · reported by psirt@cisco.comHigh 7.5
A vulnerability in the XAR file format parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition or possibly other expanded impacts as a result of memory corruption on an affected device. This vulnerability is due to improper boundary checks for content in XAR files during scanning. An attacker could exploit this vulnerability by submitting a crafted file that contains XAR content to be scanned by ClamAV on an affected device. A successful exploit could allow the attacker to cause the ClamAV scanning process to terminate, resulting in a DoS condition on the affected software.
- Product
- Cisco Secure Endpoint
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordsec.cloudapps.cisco.com
Aug 7, 2026A vulnerability in the Mach-O file format parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition or possibly other expanded iCVE-2026-20347 · Network & remote access · Cisco · reported by psirt@cisco.comHigh 7.5
A vulnerability in the Mach-O file format parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition or possibly other expanded impacts as a result of memory corruption on an affected device. This vulnerability is due to improper boundary checks for content in Mach-O files during scanning, which may result in an out-of-bounds buffer read. An attacker could exploit this vulnerability by submitting a crafted Mach-O file to be scanned by ClamAV on an affected device. A successful exploit could allow the attacker to cause the ClamAV scanning process to terminate, resulting in a DoS condition on the affected software.
- Product
- Cisco Secure Endpoint
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordsec.cloudapps.cisco.com
Aug 7, 2026A vulnerability in the PDF file format parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition or possibly other expanded impaCVE-2026-20346 · Network & remote access · Cisco · reported by psirt@cisco.comHigh 7.5
A vulnerability in the PDF file format parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition or possibly other expanded impacts as a result of memory corruption on an affected device. This vulnerability is due to improper boundary checks for content in PDF files during scanning, which may result in an out-of-bounds buffer read. An attacker could exploit this vulnerability by submitting a crafted PDF file to be scanned by ClamAV on an affected device. A successful exploit could allow the attacker to cause the ClamAV scanning process to terminate, resulting in a DoS condition on the affected software.
- Product
- Cisco Secure Endpoint
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordsec.cloudapps.cisco.com
Aug 7, 2026A vulnerability in the GPT file format parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition or possibly other expanded impaCVE-2026-20345 · Network & remote access · Cisco · reported by psirt@cisco.comHigh 7.5
A vulnerability in the GPT file format parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition or possibly other expanded impacts as a result of memory corruption on an affected device. This vulnerability is due to improper handling of an endian conversion operation, which may result in an out-of-bounds buffer write. An attacker could exploit this vulnerability by submitting a crafted GPT file to be scanned by ClamAV on an affected device. A successful exploit could allow the attacker to cause the ClamAV scanning process to terminate, resulting in a DoS condition on the affected software.
- Product
- Cisco Secure Endpoint
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordsec.cloudapps.cisco.com
Aug 7, 2026A vulnerability in the PESpin file format parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition or possibly other expanded iCVE-2026-20339 · Network & remote access · Cisco · reported by psirt@cisco.comHigh 7.5
A vulnerability in the PESpin file format parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition or possibly other expanded impacts as a result of memory corruption on an affected device. This vulnerability is due to improper boundary checks for content in PESpin files during scanning, which may result in an integer overflow. An attacker could exploit this vulnerability by submitting a crafted file that contains PESpin content to be scanned by ClamAV on an affected device. A successful exploit could allow the attacker to cause the ClamAV scanning process to terminate, resulting in a DoS condition on the affected software.
- Product
- Cisco Secure Endpoint
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordsec.cloudapps.cisco.com
Aug 7, 2026A vulnerability in the zip archive parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition on an affected device.CVE-2026-20338 · Network & remote access · Cisco · reported by psirt@cisco.comHigh 7.5
A vulnerability in the zip archive parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition on an affected device. This vulnerability is due to improper memory handling when processing content in zip files during scanning. An attacker could exploit this vulnerability by submitting a crafted zip file for scanning. A successful exploit could allow the attacker to cause the ClamAV scanning process to terminate as a result of a memory double-free, resulting in a DoS condition on the affected software.
- Product
- Cisco Secure Endpoint
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordsec.cloudapps.cisco.com
Aug 7, 2026A vulnerability in the zip archive parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition on an affected device.CVE-2026-20337 · Network & remote access · Cisco · reported by psirt@cisco.comHigh 7.5
A vulnerability in the zip archive parser of ClamAV could allow an unauthenticated, remote attacker to cause a DoS condition on an affected device. This vulnerability is due to improper boundary checks for content in zip files during scanning, which may result in an out-of-bounds write condition. An attacker could exploit this vulnerability by submitting a crafted zip file for scanning. A successful exploit could allow the attacker to cause the ClamAV scanning process to terminate, resulting in a DoS condition on the affected software.
- Product
- Cisco Secure Endpoint
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordsec.cloudapps.cisco.com
Aug 6, 2026llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1conteCVE-2026-70640 · Other software · ggml-org · reported by VulnCheckHigh 7.3
llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context. Attackers can exploit this by performing heap spray with attacker-controlled data containing a fake vtable to hijack the vtable pointer at offset +0x30, causing llama_batch_allocr::clear() to dereference arbitrary memory and achieve remote code execution.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 6, 2026llama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI wrapper where the bench_1model() function fails tCVE-2026-70639 · Other software · ggml-org · reported by VulnCheckMedium 6.8
llama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI wrapper where the bench_1model() function fails to validate the model context pointer before dereferencing it. Attackers can supply a malicious, corrupt, or truncated model file to trigger a null context condition, causing a SIGSEGV crash that terminates the Android application process and results in denial of service.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 6, 2026llama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new_1batch() function multiplies sizeoCVE-2026-70638 · Other software · ggml-org · reported by VulnCheckHigh 8.5
llama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new_1batch() function multiplies sizeof(llama_seq_id) by an attacker-controlled n_seq_max parameter without overflow validation, causing heap buffer allocation to wrap and allocate insufficient memory. Attackers can exploit this by providing a crafted n_seq_max value through a malicious model file or JNI call to trigger heap corruption and achieve denial of service or arbitrary code execution on Android applications using the LLaMA-Android binding.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 6, 2026llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokeCVE-2026-43632 · Other software · ggml-org · reported by VulnCheckCritical 9.2
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task queue and access ctx_server.vocab directly on HTTP worker threads. Attackers can exploit a time-of-check-time-of-use race condition where the main thread destroys and frees vocab after the synchronization lock is released but before the handler finishes using it, causing a crash or potential code execution when --sleep-idle-seconds is configured.
- Product
- llama.cpp
- 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
Aug 6, 2026llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds featuCVE-2026-43631 · Other software · ggml-org · reported by VulnCheckCritical 9.2
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled, allowing unauthenticated remote attackers to execute arbitrary code. Attackers can trigger the vulnerability by sending requests to affected endpoints while the server transitions to sleep mode, causing concurrent worker threads to dereference a freed vocab pointer that can be reclaimed with attacker-controlled data to achieve remote code execution.
- Product
- llama.cpp
- 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
Aug 6, 2026llama.cpp builds b4882 before b8487 contain an out-of-bounds read vulnerability in the recurrent memory state restore path that allows attackers with write acceCVE-2026-43630 · Other software · ggml-org · reported by VulnCheckMedium 6.3
llama.cpp builds b4882 before b8487 contain an out-of-bounds read vulnerability in the recurrent memory state restore path that allows attackers with write access to the slot save directory to read memory past the end of the allocated cells array. Attackers can craft a malicious slot file with an oversized seq_id value to trigger an out-of-bounds read that leaks heap data including pointer values into server logs, defeating ASLR protections and facilitating further exploitation.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 6, 2026llama.cpp builds b4882 through b9058 contain a heap buffer overflow vulnerability in the KV cache state restore path where the state_read_data() function computCVE-2026-43629 · Other software · ggml-org · reported by VulnCheckCritical 9.2
llama.cpp builds b4882 through b9058 contain a heap buffer overflow vulnerability in the KV cache state restore path where the state_read_data() function computes write size without overflow checking, allowing attackers with write access to the slot_save_path directory to corrupt heap memory. Attackers can craft malicious state files where cell_count multiplication overflows or exceeds tensor buffer allocation to write attacker-controlled bytes past buffer boundaries, potentially resulting in heap metadata corruption, model weight corruption, or arbitrary code execution via function pointer overwrite.
- Product
- llama.cpp
- 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
Aug 6, 2026llama.cpp builds b3978 through b9058 contain an integer underflow and out-of-bounds read vulnerability in the DRY sampler that allows unauthenticated attackers CVE-2026-43628 · Other software · ggml-org · reported by VulnCheckHigh 8.5
llama.cpp builds b3978 through b9058 contain an integer underflow and out-of-bounds read vulnerability in the DRY sampler that allows unauthenticated attackers to trigger a heap buffer underflow by sending a crafted HTTP request with dry_allowed_length set to INT32_MIN to the /v1/completions or /v1/chat/completions endpoints. Attackers can exploit this vulnerability to crash the server with SIGSEGV causing denial of service for all connected users, or corrupt token sampling probabilities by reading garbage values from memory before the allocated buffer.
- Product
- llama.cpp
- 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
Aug 6, 2026llama.cpp builds b1283 through b9058 contain an integer overflow vulnerability in the llama_batch_init() function where unchecked multiplications in malloc() caCVE-2026-43627 · Other software · ggml-org · reported by VulnCheckHigh 8.5
llama.cpp builds b1283 through b9058 contain an integer overflow vulnerability in the llama_batch_init() function where unchecked multiplications in malloc() calls can wrap past INT32_MAX when computing allocation sizes. Attackers can pass specially crafted parameters to trigger integer overflow, causing heap corruption and potentially achieving arbitrary code execution through subsequent batch operations that write past allocated buffer boundaries.
- Product
- llama.cpp
- 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
Aug 6, 2026llama.cpp builds b1886 through b7445 contain a double free vulnerability in the LLaMA-Android JNI wrapper where new_1batch() allocates memory using malloc() whiCVE-2026-43622 · Other software · ggml-org · reported by VulnCheckHigh 8.5
llama.cpp builds b1886 through b7445 contain a double free vulnerability in the LLaMA-Android JNI wrapper where new_1batch() allocates memory using malloc() while free_1batch() deallocates it using the C++ delete operator, causing heap metadata corruption. Attackers can trigger this memory management mismatch to cause denial of service through process crashes or potentially achieve arbitrary code execution depending on allocator state.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 5, 2026Uptime Kuma's Matomo analytics integration (server/analytics/matomo-analytics.js) injects the admin-configurable Matomo value as a bare, unquoted JavaScript expCVE-2026-71285 · Other software · louislam · reported by 309f9ea4-e3e9-4c6c-b79d-e8eb01244f2cHigh 8.1
Uptime Kuma's Matomo analytics integration (server/analytics/matomo-analytics.js) injects the admin-configurable Matomo value as a bare, unquoted JavaScript expression inside a <script> block rendered on every public status page. A siteId value such as , once saved by an editor/admin, executes arbitrary JavaScript for every unauthenticated visitor of the public /status/<slug> page, enabling session-cookie theft and full page takeover.
- Product
- uptime-kuma
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 5, 2026The raster Rust crate's crop function (src/editor.rs) clamps the crop width/height against source dimensions but only clamps the offset_x/offset_y parameters agCVE-2026-71202 · Other software · kosinix · reported by 309f9ea4-e3e9-4c6c-b79d-e8eb01244f2cHigh 7.5
The raster Rust crate's crop function (src/editor.rs) clamps the crop width/height against source dimensions but only clamps the offset_x/offset_y parameters against 0, never against the source width/height.
- Product
- raster
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 4, 2026LAMP Rapid Development Platform through 5.6.2, fixed in commit 84b0c27, contains a remote code execution vulnerability in GlueFactory that executes unsandboxed CVE-2026-69100 · Other software · dromara · reported by VulnCheckHigh 8.7
LAMP Rapid Development Platform through 5.6.2, fixed in commit 84b0c27, contains a remote code execution vulnerability in GlueFactory that executes unsandboxed Groovy scripts from database template fields without compilation restrictions or whitelisting. Attackers can write or influence the script field via message template endpoints to execute arbitrary Groovy code and OS commands on the backend server.
- Product
- lamp-cloud
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Aug 3, 2026@fastify/aws-lambda version 6.4.0 decorates each Fastify request with request.awsLambda.event and request.awsLambda.context, values that applications are documeCVE-2026-18248 · Other software · @fastify/aws-lambda · reported by ce714d77-add3-4f53-aff5-83d477b104bbCritical 9.1
@fastify/aws-lambda version 6.4.0 decorates each Fastify request with request.awsLambda.event and request.awsLambda.context, values that applications are documented to use for authorization decisions such as reading API Gateway authorizer claims. In the default configuration, the getter that populates this decoration reads the client-controlled x-apigateway-event and x-apigateway-context HTTP headers before falling back to the trusted internal request token, and those reserved headers are not stripped from the incoming event. An unauthenticated attacker who can set a single HTTP header can therefore forge the entire Lambda proxy event, including the authorizer context, and override the genuine one. This results in a full authentication and authorization bypass and privilege escalation for any application that trusts request.awsLambda.event for identity or access control. Only version 6.4.0 is affected. Patches: upgrade to @fastify/aws-lambda 6.4.1, which resolves the decoration only through the internal per-invocation token and strips the reserved headers before the request is processed.
- Product
- @fastify/aws-lambda
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordcna.openjsf.orggithub.com
Aug 2, 2026A vulnerability was determined in ggml-org llama.cpp e15efe0.CVE-2026-18581 · Other software · ggml-org · reported by VulDBLow 1.9
A vulnerability was determined in ggml-org llama.cpp e15efe0. Affected by this issue is some unknown functionality of the file common/jinja/parser.cpp of the component Jinja Minja Template Parser. Executing a manipulation with the input {{9|9|{ can lead to reachable assertion. The attack requires local access. The exploit has been publicly disclosed and may be utilized. The project was informed of the problem early through an issue report but has not responded yet.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recorddrive.proton.megithub.com
Jul 26, 2026A flaw has been found in ggml-org llama.cpp e15efe0.CVE-2026-17501 · Other software · ggml-org · reported by VulDBMedium 6.9
A flaw has been found in ggml-org llama.cpp e15efe0. This vulnerability affects the function transform of the file common/json-schema-to-grammar.cpp of the component JSON-Schema-to-GBNF Conversion. This manipulation causes uncontrolled recursion. The attack may be initiated remotely. The pull request to fix this issue awaits acceptance.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Jul 26, 2026A vulnerability was detected in ggml-org llama.cpp d006858/e15efe0.CVE-2026-17500 · Other software · ggml-org · reported by VulDBMedium 6.9
A vulnerability was detected in ggml-org llama.cpp d006858/e15efe0. This affects the function _visit_pattern of the file common/json-schema-to-grammar.cpp. The manipulation results in null pointer dereference. The attack can be launched remotely. The pull request to fix this issue awaits acceptance.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Jul 25, 2026In the Linux kernel, the following vulnerability has been resolved: iio: accel: bmc150: clamp the device-reported FIFO frame count __bmc150_accel_fifo_flush()CVE-2026-64504 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67
In the Linux kernel, the following vulnerability has been resolved: iio: accel: bmc150: clamp the device-reported FIFO frame count __bmc150_accel_fifo_flush() copies the number of samples the device reports in its hardware FIFO into an on-stack buffer u16 buffer[BMC150_ACCEL_FIFO_LENGTH * 3]; which is sized for at most BMC150_ACCEL_FIFO_LENGTH (32) samples. The frame count is read from the FIFO_STATUS register and only masked to its 7 valid bits: count = val & 0x7F; so it can be 0..127. The only other limit applied to it is the optional caller-supplied sample budget: if (samples && count > samples) count = samples; which does not constrain count on the flush-all path (samples == 0), and leaves it well above 32 whenever samples is larger. count samples are then transferred into buffer[]: bmc150_accel_fifo_transfer(data, (u8 *)buffer, count); bmc150_accel_fifo_transfer() reads count * 6 bytes through regmap, so a malfunctioning, malicious or counterfeit accelerometer (or an attacker tampering with the I2C/SPI bus) that reports up to 127 frames writes up to 762 bytes into the 192-byte buffer: a stack out-of-bounds write of up to 570 bytes that clobbers the stack canary
- 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
Jul 25, 2026In the Linux kernel, the following vulnerability has been resolved: hwrng: virtio: clamp device-reported used.len at copy_data() random_recv_done() stores theCVE-2026-64456 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67High 7.7
In the Linux kernel, the following vulnerability has been resolved: hwrng: virtio: clamp device-reported used.len at copy_data() random_recv_done() stores the device-reported used.len directly into vi->data_avail. copy_data() then indexes vi->data[] using vi->data_idx (advanced by previous copy_data() calls) and issues a memcpy() without re-validating either value against the posted buffer size sizeof(vi->data) (SMP_CACHE_BYTES bytes, typically 32 or 64). A malicious or buggy virtio-rng backend can set used.len beyond sizeof(vi->data), steering the memcpy() past the end of the inline array into adjacent kmalloc-1k slab bytes. hwrng_fillfn() mixes those bytes into the guest RNG, and guest root can also observe them directly via /dev/hwrng. Concrete impact is inside the guest: - Memory-safety / hardening: any virtio-rng backend that over-reports used.len causes the driver to read past vi->data into unrelated slab contents. hwrng_fillfn() is a kernel thread that runs as soon as the device is probed; no guest userspace interaction is required to first-trigger the OOB. - Cross-boundary leak (confidential-compute threat model): a malicious hypervisor cooperating
- 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
Jul 25, 2026In the Linux kernel, the following vulnerability has been resolved: Input: goodix - clamp the device-reported contact count goodix_ts_read_input_report() copiCVE-2026-64274 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67High 7.8
In the Linux kernel, the following vulnerability has been resolved: Input: goodix - clamp the device-reported contact count goodix_ts_read_input_report() copies the number of touch points reported by the device into an on-stack buffer u8 point_data[2 + GOODIX_MAX_CONTACT_SIZE * GOODIX_MAX_CONTACTS]; which is sized for at most GOODIX_MAX_CONTACTS (10) contacts. The only runtime check bounds the per-interrupt count against ts->max_touch_num, but that value is taken verbatim from a 4-bit field of the device configuration block and is never clamped: ts->max_touch_num = ts->config[MAX_CONTACTS_LOC] & 0x0f; The nibble can be 0..15, so a malfunctioning, malicious or counterfeit controller (or an attacker tampering with the I2C bus) can advertise up to 15 contacts. goodix_ts_read_input_report() then accepts a touch_num of up to 15 and the second goodix_i2c_read() writes ts->contact_size * (touch_num - 1) bytes past the one-contact header into point_data - up to 30 bytes (45 with the 9-byte report format) beyond the 92-byte buffer: a stack out-of-bounds write. Clamp max_touch_num to GOODIX_MAX_CONTACTS, the number of contacts point_data[] is sized for, when reading it from the conf
- 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
Jul 21, 2026Ollama (HEAD f0078ae) contains an uncontrolled memory allocation vulnerability in the GGUF metadata parser that allows remote attackers to crash the server by sCVE-2026-65315 · Other software · Ollama · reported by VulnCheckHigh 8.7
Ollama (HEAD f0078ae) contains an uncontrolled memory allocation vulnerability in the GGUF metadata parser that allows remote attackers to crash the server by supplying a crafted GGUF file with attacker-controlled length and count fields in string lengths, tensor dimension counts, and metadata array counts that are used as allocation sizes without validation against remaining file size. Attackers can upload a sub-1KB crafted GGUF file via the blob upload and model create or pull API endpoints to trigger unrecoverable Go runtime out-of-memory fatal errors or makeslice panics that bypass recovery middleware and crash the entire server process.
- Product
- Ollama
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Jul 21, 2026Improper neutralization of special elements used in an SQL command ('SQL injection') vulnerability in Turkmesh Communication Services Inc.CVE-2026-1617 · Other software · Turkmesh Communication Services Inc. · reported by iletisim@usom.gov.trCritical 9.8
Improper neutralization of special elements used in an SQL command ('SQL injection') vulnerability in Turkmesh Communication Services Inc. Turkhotspot 5651 Loglama allows SQL Injection. This issue affects Turkhotspot 5651 Loglama: from 5.1.2 before 5.1.3.
- Product
- Turkhotspot 5651 Loglama
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Jul 21, 2026Deserialization of untrusted data vulnerability that may allow class-registration checks to be bypassed during Java lambda deserialization.CVE-2026-64606 · Websites & servers · Apache Software Foundation · reported by security@apache.orgCritical 9.8
Deserialization of untrusted data vulnerability that may allow class-registration checks to be bypassed during Java lambda deserialization. Only lambda capture class is affected This issue affects Apache Fory: from before 1.4.0. Users are recommended to upgrade to version 1.4.0, which fixes the issue.
- Product
- Apache Fory
- 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
Jul 19, 2026In the Linux kernel, the following vulnerability has been resolved: tcp: fix stale per-CPU tcp_tw_isn leak enabling ISN prediction Blamed commit moved the TIMCVE-2026-64024 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67Critical 9.4
In the Linux kernel, the following vulnerability has been resolved: tcp: fix stale per-CPU tcp_tw_isn leak enabling ISN prediction Blamed commit moved the TIME_WAIT-derived ISN from the skb control block to a per-CPU variable, assuming the value would always be consumed by tcp_conn_request() for the same packet that wrote it. That assumption is violated by multiple drop paths between the producer (__this_cpu_write(tcp_tw_isn, isn) in tcp_v{4,6}_rcv()) and the consumer (tcp_conn_request()): - min_ttl / min_hopcount check - xfrm policy check - tcp_inbound_hash() MD5/AO mismatch - tcp_filter() eBPF/SO_ATTACH_FILTER drop - th->syn && th->fin discard in tcp_rcv_state_process() TCP_LISTEN - psp_sk_rx_policy_check() in tcp_v{4,6}_do_rcv() - tcp_checksum_complete() in tcp_v{4,6}_do_rcv() - tcp_v{4,6}_cookie_check() returning NULL When a packet is dropped on any of these paths, tcp_tw_isn is left set. The next SYN processed on the same CPU then consumes the non zero value in tcp_conn_request(), receiving a potentially predictable ISN. This patch moves back tcp_tw_isn to skb->cb[], getting rid of the per-cpu variable. Note that tcp_v{4,6}_fill_cb() do not set it. Very litle i
- 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
Jul 19, 2026In the Linux kernel, the following vulnerability has been resolved: Input: usbtouchscreen - clamp NEXIO data_len/x_len to URB buffer size nexio_read_data() puCVE-2026-64014 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67High 7.1
In the Linux kernel, the following vulnerability has been resolved: Input: usbtouchscreen - clamp NEXIO data_len/x_len to URB buffer size nexio_read_data() pulls data_len and x_len from a packed __be16 header in the device's interrupt packet and then walks packet->data[0..x_len) and packet->data[x_len..data_len) comparing each byte against a threshold. Both fields are 16-bit on the wire (max 65535). The existing adjustments shave at most 0x100 / 0x80 off, so the loop bound can still reach roughly 0xfeff. The URB transfer buffer for NEXIO is rept_size (1024) bytes from usb_alloc_coherent(), with the first 7 occupied by the packed header — so packet->data[] has 1017 valid bytes. read_data() callbacks are not given urb->actual_length, and nothing else bounds the walk. A device that lies about its length can get a ~64 KiB out-of-bounds read past the coherent DMA allocation. The first index whose byte exceeds NEXIO_THRESHOLD lands in begin_x / begin_y and from there into the reported touch coordinates, so adjacent kernel memory contents leak to userspace as ABS_X / ABS_Y events. Far enough out, the read can also hit an unmapped page and fault. Fix this all by clamping data_len
- 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
Jul 15, 2026Flameshot is powerful yet simple to use screenshot software.CVE-2026-62294 · Other software · flameshot-org · reported by GitHubMedium 5.1
Flameshot is powerful yet simple to use screenshot software. Prior to 14.0.0, the Open With feature wrote screenshots to a predictable temporary path and followed symlinks, creating a time-of-check to time-of-use race that allowed a local unprivileged attacker on the same machine to pre-plant a symlink and cause Flameshot to write PNG data through it, overwriting any file the victim user could write. This issue is fixed in version 14.0.0.
- Product
- flameshot
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Jul 15, 2026Quotes llama: Does not properly sanitize and escape a user-supplied parameter before using it in a SQL queryCVE-2026-12512 · WordPress plugins & themes · reported by WPScanHigh 8.6
The Quotes llama WordPress plugin before 3.1.6 does not properly sanitize and escape a user-supplied parameter before using it in a SQL query, allowing unauthenticated attackers to perform UNION-based SQL injection and read arbitrary data from the database, including password hashes.
- Product
- Quotes llama
- Fixed in
- 3.1.6
- What to do
- Update Quotes llama to 3.1.6 or newer.
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.