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 1 of 15
Oct 29, 2026Crawl4AI before 0.9.5 contains a denial of service vulnerability in the Docker API server where _clamp_untrusted() fails to cap the delay_before_return_html, meCVE-2026-119119 · Other software · unclecode · reported by VulnCheckHigh 8.7
Crawl4AI before 0.9.5 contains a denial of service vulnerability in the Docker API server where _clamp_untrusted() fails to cap the delay_before_return_html, mean_delay, and max_range fields of untrusted CrawlerRunConfig bodies. Unauthenticated attackers can send huge values to /crawl or /crawl/stream to hold browser pages or stall the rate limiter, exhausting the page pool and blocking other crawls.
- Product
- crawl4ai
- 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
Oct 29, 2026llama.cpp b2877 before b11935 contains an uncaught exception vulnerability in the ggml-rpc server that allows unauthenticated remote attackers to crash the rpc-CVE-2026-119052 · Other software · ggml-org · reported by VulnCheckHigh 8.7
llama.cpp b2877 before b11935 contains an uncaught exception vulnerability in the ggml-rpc server that allows unauthenticated remote attackers to crash the rpc-server by sending oversized message lengths. After the HELLO handshake, attackers can send a command such as SET_TENSOR with a length above std::vector::max_size(), triggering an uncaught std::length_error that aborts the process and denies service to all clients.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 27, 2026llama.cpp b8841 before b11865 contains an out-of-bounds write vulnerability in the ggml-rpc RDMA transport rdma_recv() function that allows unauthenticated peerCVE-2026-118375 · Other software · ggml-org · reported by VulnCheckCritical 9.2
llama.cpp b8841 before b11865 contains an out-of-bounds write vulnerability in the ggml-rpc RDMA transport rdma_recv() function that allows unauthenticated peers to overflow the rpc-server stack. Attackers advertising RDMA capabilities in the HELLO message can send up to 256 KiB into a 1-byte stack variable in rpc_serve_client(), crashing the server or potentially executing code.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 27, 2026rtl_433 through 25.12 contains a stack-based buffer overflow vulnerability in proflame2_mc() that allows nearby attackers to write past a 7-byte stack buffer byCVE-2026-117609 · Other software · merbanan · reported by VulnCheckHigh 7.1
rtl_433 through 25.12 contains a stack-based buffer overflow vulnerability in proflame2_mc() that allows nearby attackers to write past a 7-byte stack buffer by sending more than seven valid Manchester-coded words. Attackers can transmit a crafted OOK radio signal or supply a crafted signal file to crash rtl_433 and corrupt stack memory, enabling control-flow hijacking on builds without stack canaries.
- Product
- rtl_433
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 27, 2026Nuxt Ollama integrates the official Ollama JavaScript library into Nuxt applications.CVE-2026-59158 · Other software · thoda-dev · reported by GitHubHigh 7.5
Nuxt Ollama integrates the official Ollama JavaScript library into Nuxt applications. Prior to 1.3.1, `src/module.ts` unconditionally merges all module options, including a configured cloud `api_key`, into `runtimeConfig.public.ollama`. Nuxt serializes this public configuration into the server-rendered HTML `window.__NUXT__` payload, exposing the API key in plaintext to any unauthenticated client that fetches a page. `src/runtime/composables/useOllama.ts` also reads the public key and sends it as an `Authorization` `Bearer` credential in browser-side Ollama requests. Following the documented cloud configuration is sufficient; the visible application template need not make an Ollama request for the secret to leak. A stolen key permits unauthorized cloud API requests at the operator's expense, quota or rate-limit exhaustion, and access to stored models or conversations if the key grants those permissions. This issue is fixed in version 1.3.1.
- Product
- nuxt-ollama
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 26, 2026llama.cpp b5854 before b11275 contains an integer overflow vulnerability in gguf_init_from_reader that bypasses the CVE-2025-53630 size check when padding near-CVE-2026-116840 · Other software · ggml-org · reported by VulnCheckMedium 6.9
llama.cpp b5854 before b11275 contains an integer overflow vulnerability in gguf_init_from_reader that bypasses the CVE-2025-53630 size check when padding near-SIZE_MAX tensor sizes. Attackers can supply a crafted GGUF file whose tensor size pads to zero, causing an undersized data blob and out-of-bounds reads by consumers.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 26, 2026llama.cpp b10398 before b11301 contains an integer overflow vulnerability in gguf_init_from_reader that allows attackers to bypass tensor element-count validatiCVE-2026-116829 · Other software · ggml-org · reported by VulnCheckMedium 6.9
llama.cpp b10398 before b11301 contains an integer overflow vulnerability in gguf_init_from_reader that allows attackers to bypass tensor element-count validation by supplying shapes whose product wraps to zero. Attackers can craft GGUF files with enormous tensor extents that overflow ggml_nbytes() and strides, causing out-of-bounds memory access when buffers are sized from wrapped byte counts.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 26, 2026llama.cpp b5899 before b10261 contains a heap out-of-bounds write vulnerability in llm_tokenizer_plamo2::build that allows attackers to corrupt heap memory via CVE-2026-116816 · Other software · ggml-org · reported by VulnCheckHigh 8.5
llama.cpp b5899 before b10261 contains a heap out-of-bounds write vulnerability in llm_tokenizer_plamo2::build that allows attackers to corrupt heap memory via crafted GGUF models. Attackers can supply BYTE-typed token text like <0x-1> whose negative parsed value writes token ids before the bytes_ buffer during model loading, causing crashes or potential code execution.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 26, 2026A security vulnerability has been detected in DAMO Academy VideoLLaMA2 up to 1.1.CVE-2026-116608 · Other software · DAMO Academy · reported by VulDBLow 2.1
A security vulnerability has been detected in DAMO Academy VideoLLaMA2 up to 1.1. Affected by this issue is the function load_pretrained_model of the file videollama2/model/__init__.py of the component Model Loading. The manipulation of the argument _name_or_path leads to improper validation of integrity check value. Remote exploitation of the attack is possible. The exploit has been disclosed publicly and may be used. The project was informed of the problem early through an issue report but has not responded yet.
- Product
- VideoLLaMA2
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 26, 2026llama.cpp before b5649 contains a heap out-of-bounds read vulnerability in llama_vocab::impl::print_info() that allows attackers to crash model loading via overCVE-2025-71439 · Other software · ggml-org · reported by VulnCheckMedium 6.8
llama.cpp before b5649 contains a heap out-of-bounds read vulnerability in llama_vocab::impl::print_info() that allows attackers to crash model loading via oversized special token IDs. Attackers can supply a crafted GGUF model with a short tokenizer.ggml.tokens array, causing llama-cli or llama-server to read past id_to_token and crash.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 26, 2026llama.cpp before b3701 contains an out-of-bounds read vulnerability in llama_decode_internal and llama_encode_internal that allows remote unauthenticated attackCVE-2024-58400 · Other software · ggml-org · reported by VulnCheckHigh 8.8
llama.cpp before b3701 contains an out-of-bounds read vulnerability in llama_decode_internal and llama_encode_internal that allows remote unauthenticated attackers to read memory by supplying unvalidated token IDs. Attackers can send integer-array prompts to the llama-server /completion endpoint with IDs outside 0 to n_vocab to read embedding memory out of bounds or crash the process.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 26, 2026llama.cpp before b2731 contains an out-of-bounds read vulnerability in llama_model_loader that allows attackers to read memory outside the mapped model file viaCVE-2024-58399 · Other software · ggml-org · reported by VulnCheckMedium 6.9
llama.cpp before b2731 contains an out-of-bounds read vulnerability in llama_model_loader that allows attackers to read memory outside the mapped model file via crafted GGUF tensor offsets. Attackers can supply a crafted or truncated GGUF model file with arbitrary 64-bit tensor offsets to crash the process or load adjacent process memory as model weights.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 26, 2026A flaw has been found in turboderp-org ExLlamaV2 up to 0.3.2.CVE-2026-116502 · Other software · turboderp-org · reported by VulDBMedium 4.8
A flaw has been found in turboderp-org ExLlamaV2 up to 0.3.2. The impacted element is the function QMatrix::make_sequential of the file exllamav2/exllamav2_ext/cuda/q_matrix.cu of the component GPTQ Index Handler. Executing a manipulation of the argument g_idx can lead to out-of-bounds write. The attack needs to be launched locally. The project was informed of the problem early through an issue report but has not responded yet.
- Product
- ExLlamaV2
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 26, 2026A security flaw has been discovered in ggml-org llama.cpp up to 0.7.0.CVE-2026-116467 · Other software · ggml-org · reported by VulDBMedium 6.9
A security flaw has been discovered in ggml-org llama.cpp up to 0.7.0. Affected by this vulnerability is an unknown functionality of the file src/llama-sampler.cpp of the component Mirostat Sampler. The manipulation results in reachable assertion. It is possible to launch the attack remotely. The project was informed of the problem early through a pull request but has not reacted yet.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 25, 2026A weakness has been identified in ggml-org llama.cpp.CVE-2026-116096 · Other software · ggml-org · reported by VulDBLow 1.9
A weakness has been identified in ggml-org llama.cpp. Affected by this issue is the function list_pockettts_state_slots of the file tools/mtmd/models/pockettts-gen.cpp of the component Pocket-TTS. This manipulation of the argument mmproj causes out-of-bounds write. The attack requires local access. The exploit has been made available to the public and could be used for attacks. Upgrading to version b11796 can resolve this issue. Patch name: 0a1ef309cdbb9f2024f58a38fb4c42352c5db569. The affected component should be upgraded. The reported heap out‑of‑bounds write (gen‑audio / pocket‑tts state_out path) is no longer a silent exploitable heap write on current master and the latest release b11871 - a malicious mmproj that previously triggered it now trips a release‑active GGML_ASSERT and aborts. But the submitter's recommended by‑construction fix (PR #29360 "prevent OOB accesses from malformed mmproj files") is still open. So the root cause is detected‑and‑aborted, not prevented‑by‑design.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 23, 2026The RTIO streaming completion callback adxl345_process_fifo_samples_cb() in drivers/sensor/adi/adxl345/adxl345_stream.c clamps the amount of FIFO data it will rCVE-2026-79978 · Other software · zephyrproject · reported by vulnerabilities@zephyrproject.orgLow 3.6
The RTIO streaming completion callback adxl345_process_fifo_samples_cb() in drivers/sensor/adi/adxl345/adxl345_stream.c clamps the amount of FIFO data it will report (read_len = MIN(fifo_bytes, buf_avail)) to the space actually returned by rtio_sqe_rx_buf(), but then queues one 6-byte bus read per hardware FIFO entry at consecutive offsets from read_buf, bounded by the device-reported entry count rather than by the clamp. When the allocated buffer is smaller than a full batch, the driver writes up to 186 bytes past the end of the allocation; a companion defect left data->fifo_total_bytes un-reset at batch start, so an aborted or errored chain could leave a stale write offset that pushes the next batch's writes out of bounds as well. A short buffer is an expected, first-class outcome of the RTIO allocator: rtio_block_pool_alloc() walks the requested size down one block at a time until a contiguous run is available, and the driver only demands sizeof(struct adxl345_fifo_data) + 6 bytes as its minimum. Any streaming configuration whose mempool block granularity is smaller than a full FIFO batch (204 bytes) — for example the sensor shell's RTIO_DEFINE_WITH_MEMPOOL(sensor_read_rtio, 8,
- Product
- zephyr
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 19, 2026A vulnerability was found in LlamaChinese Llama-Chinese up to 6fa0fffb0dd82fe3cfaa1449ee54a5806d26ae9b.CVE-2026-112080 · Other software · LlamaChinese · reported by VulDBLow 2.1
A vulnerability was found in LlamaChinese Llama-Chinese up to 6fa0fffb0dd82fe3cfaa1449ee54a5806d26ae9b. The impacted element is the function get_class_from_dynamic_module of the file examples/chat_gradio_no_merge.py of the component Adapter Config. The manipulation of the argument base_model_name_or_path results in code injection. The attack may be performed from remote. The exploit has been made public and could be used. This product utilizes a rolling release system for continuous delivery, and as such, version information for affected or updated releases is not disclosed. The project was informed of the problem early through an issue report but has not responded yet.
- Product
- Llama-Chinese
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 18, 2026In the Linux kernel, the following vulnerability has been resolved: drm/imagination: clamp freelist reconstruction requests The firmware reconstruction count CVE-2026-98566 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67
In the Linux kernel, the following vulnerability has been resolved: drm/imagination: clamp freelist reconstruction requests The firmware reconstruction count controls accesses to the request's fixed freelist ID array and the copy into the fixed response array. Neither access currently bounds the count to those protocol arrays. Clamp the count to the request capacity, which is shared by the response layout, and use that count consistently for reconstruction and response publication. Keep the firmware recovery exchange instead of dropping an oversized request without a response, as discussed with the firmware maintainer. The issue was found by our static-analysis tool.
- 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
Oct 18, 2026Schism Tracker through 20260524 contains a heap out-of-bounds write vulnerability in fmt_ult_load_song() in fmt/ult.c that fails to clamp the one-byte sample coCVE-2026-112007 · Other software · schismtracker · reported by VulnCheckMedium 6.9
Schism Tracker through 20260524 contains a heap out-of-bounds write vulnerability in fmt_ult_load_song() in fmt/ult.c that fails to clamp the one-byte sample count to MAX_SAMPLES. Attackers can supply a crafted .ult module that overwrites the instruments pointer array, causing csf_destroy() to free attacker-chosen pointers and crash the application.
- Product
- Schism Tracker
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 16, 2026ChatOllama through commit 96c3ce5 contains an insecure direct object reference vulnerability that allows remote attackers to access other users' private knowledCVE-2026-111702 · Other software · sugarforever · reported by VulnCheckHigh 8.7
ChatOllama through commit 96c3ce5 contains an insecure direct object reference vulnerability that allows remote attackers to access other users' private knowledge bases by supplying sequential ids. Attackers can enumerate ids via GET /api/knowledgebases/{id} or the knowledgebaseId parameter of POST /api/models/chat to read metadata and retrieved document content.
- Product
- chat-ollama
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 15, 2026libmikmod through 3.3.15 contains a global out-of-bounds read vulnerability in the MultiTracker loader MTM_Load() because the module channel count is never clamCVE-2026-111165 · Other software · sezero · reported by VulnCheckMedium 6.8
libmikmod through 3.3.15 contains a global out-of-bounds read vulnerability in the MultiTracker loader MTM_Load() because the module channel count is never clamped. Attackers can supply a crafted .mtm file with a channel count above 32 to read past the static pat array, corrupting track indices and crashing the loading application.
- Product
- libmikmod
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 14, 2026FreeRDP 2.0.0 before 3.32.0, when built with WITH_LAME=ON and WITH_DSP_FFMPEG=OFF, has a heap-based buffer overflow in freerdp_dsp_decode_mp3 because the right-CVE-2026-110837 · Other software · FreeRDP · reported by VulnCheckMedium 6.1
FreeRDP 2.0.0 before 3.32.0, when built with WITH_LAME=ON and WITH_DSP_FFMPEG=OFF, has a heap-based buffer overflow in freerdp_dsp_decode_mp3 because the right-channel PCM pointer is offset by a byte count treated as short elements. A malicious RDP server can send MP3 audio over rdpsnd so decoded samples cross the heap allocation, crashing the client.
- Product
- FreeRDP
- 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
Oct 14, 2026A vulnerability was detected in turboderp ExLlamaV2 up to 0.3.2.CVE-2026-110695 · Other software · turboderp · reported by VulDBLow 2.1
A vulnerability was detected in turboderp ExLlamaV2 up to 0.3.2. Affected is the function STFile.get_tensor of the file exllamav2/stloader.py of the component Safetensors Tensor Loader. Performing a manipulation of the argument data_offsets results in buffer overflow. The attack is possible to be carried out remotely. The exploit is now public and may be used. The project was informed of the problem early through an issue report but has not responded yet.
- Product
- ExLlamaV2
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 14, 2026LDAP Account Manager (LAM) is a webfrontend for managing entries (e.g.CVE-2026-55167 · Other software · LDAPAccountManager · reported by GitHubMedium 4.2
LDAP Account Manager (LAM) is a webfrontend for managing entries (e.g. users, groups, DHCP settings) stored in an LDAP directory. Prior to 9.6, installations with WebAuthn two-factor authentication leave the challenge in the same PHP session after the authentication response is processed. A user who has already completed valid password authentication can therefore retain the same signed WebAuthn ceremony state instead of consuming it once. The practical risk is limited by the requirement for a valid password, the same PHP session, and a valid authenticator signature. This issue is fixed in version 9.6.
- Product
- lam
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 14, 2026lamp-boot 4.16.0-java17 through 4.21.2-java17 contains an insecure direct object reference vulnerability in the oauth module's /anyone/findDeptByCompany endpoinCVE-2026-110763 · Other software · zuihou · reported by VulnCheckMedium 5.3
lamp-boot 4.16.0-java17 through 4.21.2-java17 contains an insecure direct object reference vulnerability in the oauth module's /anyone/findDeptByCompany endpoint that skips URI permission checks. Authenticated low-privilege users can iterate sequential employeeId values to read every employee's department membership within the tenant.
- Product
- lamp-boot
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 14, 2026BentoPDF is a privacy-focused PDF toolkit that runs in the browser.CVE-2026-102354 · Other software · alam00000 · reported by GitHubMedium 6.8
BentoPDF is a privacy-focused PDF toolkit that runs in the browser. In versions through 2.8.6, cloudflare/cors-proxy-worker.js performs a DNS-over-HTTPS safety check in hostnameResolvesToPrivate(), as used by isValidCertificateUrl() and isValidTsaUrl(), separately from the outbound fetch(targetUrl, ...). In deployments where PROXY_SECRET validation is not enabled, a direct client that controls authoritative DNS can forge the Origin header, supply a certificate, OCSP, CRL, or TSA URL that passes the proxy's accepted path checks, return a public address during validation, and return a private or link-local address during the later fetch. The endpoint can return allowed response bodies to the attacker, but exploitation requires precise timing, redirects and IP-literal obfuscation are blocked, and network reachability limits the target. This vulnerability is fixed in 2.8.7.
- Product
- bentopdf
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 12, 2026ExLlamaV3’s EXL3 loader uses an untrusted, model-supplied index to select a CUDA kernel without first verifying that the index is within bounds.CVE-2026-84286 · Other software · Turboderp-org · reported by cret@cert.orgCritical 9.8
ExLlamaV3’s EXL3 loader uses an untrusted, model-supplied index to select a CUDA kernel without first verifying that the index is within bounds. A malicious model can force the code to read and invoke an out‑of‑range function pointer, resulting in a reproducible crash. Under specific conditions involving certain host memory layouts, it may be possible for an attacker to exploit this vulnerability to achieve controlled indirect kernel dispatch during model load.
- Product
- ExLlamaV3
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 12, 2026FreeRDP before 3.32.0 stores the server-supplied signed MS-RDPEGT geometry origin unchecked in an unsigned video surface coordinate, so the blit clamp in gdiVidCVE-2026-109251 · Other software · FreeRDP · reported by VulnCheckMedium 6
FreeRDP before 3.32.0 stores the server-supplied signed MS-RDPEGT geometry origin unchecked in an unsigned video surface coordinate, so the blit clamp in gdiVideoShowSurface() wraps in 32-bit arithmetic and is skipped. A malicious RDP server can send a negative topLevelLeft to clients connecting with /video, causing an out-of-bounds write of decoded H.264 picture data or a client abort.
- Product
- FreeRDP
- 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
Oct 12, 2026lamp-cloud through 5.10.1 contains an insecure direct object reference vulnerability that allows authenticated users to list other business objects' file attachCVE-2026-109168 · Other software · dromara · reported by VulnCheckHigh 7.1
lamp-cloud through 5.10.1 contains an insecure direct object reference vulnerability that allows authenticated users to list other business objects' file attachments by supplying an arbitrary bizId to the listFileByBizId endpoint. Attackers can retrieve file names, storage paths, MD5 hashes and stored URLs, which in default LOCAL storage are served without login under /anyTenant/file.
- Product
- lamp-cloud
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 12, 2026lamp-cloud through 5.10.1 contains an insecure direct object reference vulnerability that allows authenticated users to obtain storage URLs for other users' filCVE-2026-109166 · Other software · dromara · reported by VulnCheckHigh 7.1
lamp-cloud through 5.10.1 contains an insecure direct object reference vulnerability that allows authenticated users to obtain storage URLs for other users' files by supplying arbitrary file ids. Attackers can send another user's file ids to the findUrlById handler of FileAnyoneController, which never checks ownership, to retrieve access URLs and read uploaded attachments.
- Product
- lamp-cloud
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 12, 2026A vulnerability has been found in Ollama up to 0.34.1.CVE-2026-108836 · Other software · reported by VulDBMedium 5.5
A vulnerability has been found in Ollama up to 0.34.1. The affected element is the function IsValid of the file types/model/name.go of the component Pull API. The manipulation leads to server-side request forgery. It is possible to initiate the attack remotely. The exploit has been disclosed to the public and may be used. Upgrading to version 0.34.2 is sufficient to fix this issue. The identifier of the patch is dfabde4539e42ba1e1eab50a3a50b88aea7958a0. It is suggested to upgrade the affected component.
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 11, 2026LlamaFarm through 0.0.34 contains an insecure default configuration that binds its unauthenticated FastAPI server to 0.0.0.0 on port 14345, while the lf CLI silCVE-2026-108760 · Other software · llama-farm · reported by VulnCheckHigh 7.2
LlamaFarm through 0.0.34 contains an insecure default configuration that binds its unauthenticated FastAPI server to 0.0.0.0 on port 14345, while the lf CLI silently discards HOST overrides. Network-adjacent attackers can call the project and dataset management API to read stored provider API keys, modify projects, trigger ingestion, and irreversibly delete projects.
- Product
- llamafarm
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 8, 2026Ollama is vulnerable to path traversal in the `/api/pull` endpoint due to insufficient validation of layer digests by the `digestToPath` function.CVE-2026-103663 · Other software · Ollama · reported by cvd@cert.plCritical 9.4
Ollama is vulnerable to path traversal in the `/api/pull` endpoint due to insufficient validation of layer digests by the `digestToPath` function. An unauthenticated remote attacker can specify a path traversal sequence as a layer digest, causing a malicious binary to be written outside the model store. Critically if the server process has write access to `/usr/lib/ollama` (the default in most Ollama Docker images), an attacker can write the malicious file to that directory. On the next server restart, the file is loaded and executed, resulting in remote code execution as root. This issue was fixed in version 0.35.0.
- Product
- Ollama
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 7, 2026llama.cpp before b11393 contains a use-after-free and double free vulnerability in common_chat_peg_mapper::map that allows unauthenticated remote attackers to cCVE-2026-107183 · Other software · ggml-org · reported by VulnCheckCritical 9.2
llama.cpp before b11393 contains a use-after-free and double free vulnerability in common_chat_peg_mapper::map that allows unauthenticated remote attackers to corrupt heap memory via a dangling current_tool pointer. Attackers can submit a chat_parser in a POST /completion request emitting a tool-id after a tool-close tag to crash llama-server and shape a heap write primitive.
- Product
- llama.cpp
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Oct 6, 2026Authorization bypass through a user-controlled key in the optional Amazon Q Business Lambda hook sample ( q-business-lambda-hook https://github.com/aws-solutionCVE-2026-105811 · Other software · aws · reported by ff89ba41-3aa1-4d27-914a-91399e9639e5High 7.1
Authorization bypass through a user-controlled key in the optional Amazon Q Business Lambda hook sample ( q-business-lambda-hook https://github.com/aws-solutions-library-samples/qnabot-on-aws/blob/main/source/docs/lambda_hooks/README.md ), available with QnABot on AWS versions 7.0.0 through 7.4.5, might allow an authenticated remote user to read arbitrary Amazon S3 objects in the deploying AWS account. This sample solution provides an example Lambda hook and requires separate, manual deployment and additional setup. It is not deployed automatically with QnABot. Customers who have not deployed this optional sample hook are not affected and do not need to take action. To remediate this issue, affected customers should update the QnABot on AWS stack to version 7.4.6 or later and then redeploy the Amazon Q Business Lambda hook sample stack. Updating the QnABot on AWS stack alone does not deliver the fix.
- Product
- qnabot-on-aws
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordgithub.comstaging.prod.website.marketing.aws.dev
Oct 6, 2026In the Linux kernel, the following vulnerability has been resolved: ALSA: usb-audio: Clamp implicit feedback packet count to URB capacity data_ep_set_params()CVE-2026-98265 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67
In the Linux kernel, the following vulnerability has been resolved: ALSA: usb-audio: Clamp implicit feedback packet count to URB capacity data_ep_set_params() allocates each data URB for exactly u->packets isochronous frames, so urb->iso_frame_desc[] has u->packets slots and ctx->packets is the driver's only record of that limit. For an implicit feedback sink, snd_usb_queue_pending_output_urbs() overwrites it with the sync source's packet count, which is calculated independently from the capture endpoint's parameters. When that count is larger, prepare_playback_urb() and prepare_silent_urb() can write iso_frame_desc[] past the allocation; their existing bounds limit payload bytes, not the descriptor index. The reproducer uses a high-speed UAC2 device declaring bInterval 1 for implicit feedback capture (8 packets) and bInterval 4 for playback (1 packet). On the first capture completion after the stream starts, it accesses seven descriptors spanning 112 bytes beyond the one-packet URB: BUG: KASAN: slab-out-of-bounds in prepare_playback_urb (sound/usb/pcm.c:1560) Write of size 4 at addr ffff88801e696ad0 by task vhci_rx/178 prepare_playback_urb (sound/usb/pcm.c:1560) prep
- 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
Oct 6, 2026In the Linux kernel, the following vulnerability has been resolved: Input: cyttsp5 - clamp the HID report size before memcpy The size field comes from the devCVE-2026-98206 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67
In the Linux kernel, the following vulnerability has been resolved: Input: cyttsp5 - clamp the HID report size before memcpy The size field comes from the device and is used as the memcpy() length into response_buf, which is CY_MAX_INPUT bytes.
- 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
Oct 5, 2026Canimaan Software ClamXAV versions 3.3 - 3.11 contains a local privilege escalation vulnerability in the Privileged Helper Tool caused by a race condition and iCVE-2026-105773 · Other software · Canimaan Software · reported by 9119a7d8-5eab-497f-8521-727c672e3725High 7.3
Canimaan Software ClamXAV versions 3.3 - 3.11 contains a local privilege escalation vulnerability in the Privileged Helper Tool caused by a race condition and insufficient file validation, allowing a local attacker to execute arbitrary code with system privileges. Fixed in 3.11.1.
- Product
- ClamXAV
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordclamxav.comraw.githubusercontent.com
Oct 1, 2026A fail-open error handling issue within the data masking utility of Powertools for AWS Lambda (Python) might allow actors to read sensitive field values that thCVE-2026-104002 · Other software · AWS · reported by ff89ba41-3aa1-4d27-914a-91399e9639e5Medium 6
A fail-open error handling issue within the data masking utility of Powertools for AWS Lambda (Python) might allow actors to read sensitive field values that the application intended to mask. To remediate this issue, users should upgrade to version 3.35.0.
- Product
- powertools-lambda-python
- 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
Oct 1, 2026Filament is a collection of full-stack components for accelerated Laravel development.CVE-2026-104181 · Other software · filamentphp · reported by GitHubMedium 5.4
Filament is a collection of full-stack components for accelerated Laravel development. From 4.0.0 until 4.13.3 and 5.8.3, app-based multi-factor authentication management actions do not consistently require confirmation of the current password. An attacker with access to an authenticated user session can set up app-based MFA and obtain recovery codes, or disable app-based MFA and regenerate recovery codes by supplying an existing app code or recovery code, without knowing the account password. Email-based MFA is not affected, and the issue does not independently permit an unauthenticated sign-in, but changing the app-MFA configuration may lock the legitimate user out. This issue is fixed in versions 4.13.3 and 5.8.3.
- Product
- filament
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 30, 2026Authorization bypass through User-Controlled key vulnerability in The Wikimedia Foundation MediaWiki WikiLambda extension allows Authentication Bypass.CVE-2026-103446 · Other software · The Wikimedia Foundation · reported by c4f26cc8-17ff-4c99-b5e2-38fc1793eaccHigh 7.4
Authorization bypass through User-Controlled key vulnerability in The Wikimedia Foundation MediaWiki WikiLambda extension allows Authentication Bypass. This issue affects MediaWiki WikiLambda extension: 1.46.
- Product
- MediaWiki WikiLambda extension
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordgerrit.wikimedia.orgphabricator.wikimedia.org
Sep 29, 2026Improper neutralization of input during web page generation ('cross-site scripting') vulnerability in Wikimedia Foundation MediaWiki - WikiLambda extension alloCVE-2026-103046 · Other software · The Wikimedia Foundation · reported by c4f26cc8-17ff-4c99-b5e2-38fc1793eaccMedium 6.1
Improper neutralization of input during web page generation ('cross-site scripting') vulnerability in Wikimedia Foundation MediaWiki - WikiLambda extension allows Stored XSS. This issue affects MediaWiki - WikiLambda extension: before 1.46.1.
- Product
- Mediawiki - WikiLambda Extension
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordgerrit.wikimedia.orgphabricator.wikimedia.org
Sep 29, 2026Ollama versions 0.14.0 before 0.31.2 contain an incorrect authorization vulnerability in the experimental agent mode Bash tool approval mechanism that fails to CVE-2026-102697 · Other software · ollama · reported by VulnCheckHigh 8.5
Ollama versions 0.14.0 before 0.31.2 contain an incorrect authorization vulnerability in the experimental agent mode Bash tool approval mechanism that fails to properly parse shell syntax. Attackers who can influence model output through prompt injection can execute additional shell commands by appending control operators like semicolons or logical operators to approved commands, bypassing the session approval requirement.
- Product
- ollama
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 25, 2026Flame through 2.4.0 contains an insufficient session expiration vulnerability in the login endpoint that allows attackers with former admin access to obtain tokCVE-2026-100502 · Other software · pawelmalak · reported by VulnCheckMedium 5.9
Flame through 2.4.0 contains an insufficient session expiration vulnerability in the login endpoint that allows attackers with former admin access to obtain tokens with arbitrary lifespans by supplying unvalidated duration parameters. Attackers can mint near-permanent administrator tokens that survive password changes, retaining full control of the dashboard since tokens are verified only against a static JWT secret that is never rotated.
- Product
- flame
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 25, 2026Flame through 2.4.0 contains an improper restriction of excessive authentication attempts vulnerability in the POST /api/auth login endpoint that allows unautheCVE-2026-100501 · Other software · pawelmalak · reported by VulnCheckHigh 8.3
Flame through 2.4.0 contains an improper restriction of excessive authentication attempts vulnerability in the POST /api/auth login endpoint that allows unauthenticated attackers to brute-force the admin password. Attackers can submit unlimited password guesses without rate limiting, attempt counters, lockouts, or delays to gain full administrator access and modify application configuration.
- Product
- flame
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 25, 2026Flame through 2.4.0 contains an information exposure vulnerability in the unauthenticated GET /api/config endpoint that returns the entire configuration object CVE-2026-100418 · Other software · pawelmalak · reported by VulnCheckMedium 6.9
Flame through 2.4.0 contains an information exposure vulnerability in the unauthenticated GET /api/config endpoint that returns the entire configuration object without field redaction. Attackers can retrieve the stored weather API key and internal operational settings by sending a single unauthenticated request to consume provider quota or access sensitive configuration data.
- Product
- flame
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Sep 25, 2026Improper Neutralization of Input During Web Page Generation (XSS or 'Cross-site Scripting') vulnerability in Wikimedia Foundation Mediawiki - WikiLambda ExtensiCVE-2026-100383 · Other software · Wikimedia Foundation · reported by c4f26cc8-17ff-4c99-b5e2-38fc1793eaccMedium 4.8
Improper Neutralization of Input During Web Page Generation (XSS or 'Cross-site Scripting') vulnerability in Wikimedia Foundation Mediawiki - WikiLambda Extension allows Cross-Site Scripting (XSS). This issue affects Mediawiki - WikiLambda Extension: 1.47.0-alpha. The issue has been remediated on the `master` branch.
- Product
- Mediawiki - WikiLambda Extension
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordgerrit.wikimedia.orgphabricator.wikimedia.org
Sep 25, 2026Exposure of Sensitive Information to an Unauthorized Actor vulnerability in Wikimedia Foundation Mediawiki - WikiLambda Extension allows Excavation.CVE-2026-100377 · Other software · Wikimedia Foundation · reported by c4f26cc8-17ff-4c99-b5e2-38fc1793eaccMedium 6.9
Exposure of Sensitive Information to an Unauthorized Actor vulnerability in Wikimedia Foundation Mediawiki - WikiLambda Extension allows Excavation. This issue affects Mediawiki - WikiLambda Extension: 1.47.0-alpha. The issue has been remediated on the `master` branch.
- Product
- Mediawiki - WikiLambda Extension
- What to do
- Apply the vendor’s update. The NVD record links to the advisory.
Full record and what to do →NVD recordgerrit.wikimedia.orgphabricator.wikimedia.org
Sep 25, 2026In the Linux kernel, the following vulnerability has been resolved: net/sched: fq_pie: clamp quantum in change path fq_pie_change() accepts any quantum value CVE-2026-98013 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67
In the Linux kernel, the following vulnerability has been resolved: net/sched: fq_pie: clamp quantum in change path fq_pie_change() accepts any quantum value from userspace, including 1. With a crafted size table qdisc_pkt_len reaches ~2 GiB, so quantum=1 makes the deficit-refill loop spin ~2^31 times under the qdisc lock (a soft lockup / denial of service). Add max(256U, ...) matching fq_codel_change(). Conditions to recreate the bug: CONFIG_NET_SCH_FQ_PIE=y. Requires CAP_NET_ADMIN (namespace-local via unshare -Urn suffices). tc qdisc add dev dummy0 root fq_pie tc qdisc change dev dummy0 root fq_pie quantum 1 stab data 32768 size_log 15 cell_log 0
- 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 25, 2026In the Linux kernel, the following vulnerability has been resolved: net/sched: sfq: clamp quantum in change path sfq_change() accepts any non-negative quantumCVE-2026-98012 · Computers, phones & browsers · Linux · reported by 416baaa9-dc9f-4396-8d5f-8c081fb06d67
In the Linux kernel, the following vulnerability has been resolved: net/sched: sfq: clamp quantum in change path sfq_change() accepts any non-negative quantum (only rejects (int)ctl->quantum < 0). With a crafted size table qdisc_pkt_len reaches ~2 GiB, so quantum=1 makes the deficit-refill loop spin ~2^31 times under the qdisc lock (a soft lockup / denial of service). Add max(256U, ...) matching fq_codel_change(). Reject quantum > 1<<20 with -EINVAL, matching fq_codel_change() and the init clamp. Conditions to recreate the bug: CONFIG_NET_SCH_SFQ=y. Requires CAP_NET_ADMIN (namespace-local via unshare -Urn suffices). tc qdisc add dev dummy0 root sfq tc qdisc change dev dummy0 root sfq quantum 1 stab data 32768 size_log 15 cell_log 0
- 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
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.