News & Sicurezza

Crawl4AI before 0.8.7 contains a server-side request forgery vulnerability in the /crawl, /crawl/stream, /md, and /llm endpoints that fetch arbitrary user-supplied URLs without validation. Unauthenticated attackers can bypass the internal-address blocklist using IPv6-mapped IPv4 addresses to reach internal services and cloud metadata endpoints.

picklescan before 0.0.29 fails to detect malicious pickle files that exploit idlelib.autocomplete.AutoComplete.get_entity function in reduce methods. Attackers can embed undetected code in pickle files that executes arbitrary commands when loaded by victims using pickle.load().

picklescan before 0.0.30 (affected versions 0.0.26 and earlier) fails to detect the ensurepip._run_pip built-in function when scanning pickle files, allowing attackers to execute arbitrary code. Malicious pickle files embedding ensurepip._run_pip calls in __reduce__ methods bypass picklescan detection and achieve remote code execution upon pickle.load() invocation.

Picklescan before 0.0.33 fails to detect the numpy.f2py.crackfortran._eval_length gadget in pickle __reduce__ methods, allowing arbitrary code execution. Attackers can craft malicious pickle files that execute arbitrary Python code when loaded by victims who trust Picklescan's safety validation.

CVE ID :CVE-2026-54232 Published : June 22, 2026, 10:16 p.m. | 13 hours, 27 minutes ago Description :vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.1, the vLLM Dockerfile is vulnerable to a dependency confusion attack through the flashinfer-jit-cache package. The package is installed from a custom index (flashinfer.ai/whl/) using --extra-index-url, but the package name was not registered on PyPI, and UV_INDEX_STRATEGY="unsafe-best-match" is set globally. An attacker who registers flashinfer-jit-cache on PyPI with version 0.6.11.post2 can execute arbitrary code as root during the Docker build and backdoor every resulting container image, enabling exfiltration of all user prompts, API credentials, and model data from production vLLM deployments This vulnerability is fixed in 0.22.1. Severity: 8.8 | HIGH Visit the link for more details, such as CVSS details, affected products, timeline, and more...

CVE ID :CVE-2026-54233 Published : June 22, 2026, 10:10 p.m. | 13 hours, 33 minutes ago Description :vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.23.1rc0, vLLM's /v1/audio/transcriptions endpoint limits compressed upload size but not decoded PCM output. A 25MB OPUS file expands to ~14.9GB of float32 PCM at decode time. This vulnerability is fixed in 0.23.1rc0. Severity: 6.5 | MEDIUM Visit the link for more details, such as CVSS details, affected products, timeline, and more...

CVE ID :CVE-2026-54236 Published : June 22, 2026, 10:09 p.m. | 13 hours, 34 minutes ago Description :vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.23.1rc0, the fix for CVE-2026-22778, which introduced a sanitize_message helper that strips object-repr memory addresses from error messages before they reach the client, is incomplete: several response paths echo str(exc) directly to clients without calling sanitize_message. The unsanitized sites include the Anthropic API router in vllm/entrypoints/anthropic/api_router.py (the POST /v1/messages and POST /v1/messages/count_tokens handlers), the Server-Sent Events streaming converter in vllm/entrypoints/anthropic/serving.py, and the realtime speech-to-text WebSocket in vllm/entrypoints/speech_to_text/realtime/connection.py. These paths catch the exception inside the route coroutine and construct the JSONResponse themselves, bypassing the sanitizing global FastAPI exception handler, and WebSocket frames do not traverse that handler chain at all. Using the same primitive as the parent issue, an unauthenticated attacker can send malformed image bytes through the Anthropic Messages API image content parts so that PIL.Image.open raises an UnidentifiedImageError whose message contains the BytesIO object repr, leaking the heap memory address verbatim in the error.message field of the response body. This vulnerability is fixed in 0.23.1rc0. Severity: 5.3 | MEDIUM Visit the link for more details, such as CVSS details, affected products, timeline, and more...

CVE ID :CVE-2026-54235 Published : June 22, 2026, 9:59 p.m. | 13 hours, 45 minutes ago Description :vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.23.1rc0, ll temperature validation gates use comparison operators (), which silently evaluate to False for NaN and for positive Infinity in Python's IEEE 754 float semantics. Both values pass every guard and propagate to GPU sampling kernels, where they produce undefined behavior or CUDA errors that can crash the inference worker. This vulnerability is fixed in 0.23.1rc0. Severity: 6.9 | MEDIUM Visit the link for more details, such as CVSS details, affected products, timeline, and more...

CVE ID :CVE-2026-48746 Published : June 22, 2026, 9:57 p.m. | 13 hours, 46 minutes ago Description :vLLM is an inference and serving engine for large language models (LLMs). From 0.3.0 until 0.22.0, a vulnerability in ASGI web servers and starlette's trust on those web servers enables an authentication bypass of the OpenAI API AuthenticationMiddleware. It allows to use the API without providing the configured VLLM_API_KEY or --api-key. This vulnerability is fixed in 0.22.0. Severity: 9.1 | CRITICAL Visit the link for more details, such as CVSS details, affected products, timeline, and more...

CVE ID :CVE-2026-53923 Published : June 22, 2026, 9:55 p.m. | 13 hours, 48 minutes ago Description :vLLM is an inference and serving engine for large language models (LLMs). From 0.5.5 until 0.23.1rc0, integer truncation of tensor dimensions in vLLM's GGUF dequantize kernels (csrc/quantization/gguf/gguf_kernel.cu) causes partial tensor processing. The output tensor is allocated at full size via torch::empty (uninitialized memory), but the dequantize CUDA kernel processes only a truncated number of elements. The unfilled portion of the output tensor retains whatever was previously in GPU memory. In multi-tenant inference deployments, this residual GPU memory may contain tensor data from other users' inference requests, constituting information disclosure. This vulnerability is fixed in 0.23.1rc0. Severity: 5.3 | MEDIUM Visit the link for more details, such as CVSS details, affected products, timeline, and more...

CVE ID :CVE-2026-55409 Published : June 22, 2026, 9:47 p.m. | 13 hours, 56 minutes ago Description :Filament is a collection of full-stack components for accelerated Laravel development. From 3.0.0 until 3.3.53, a disabled RichEditor field rendered its raw state without sanitizing HTML. Where the data stored in this field's state isn't sanitized already when the form state was filled, an attacker could plant malicious HTML or JavaScript and achieve XSS that executes for users who view the form. This vulnerability is fixed in 3.3.53. Severity: 7.6 | HIGH Visit the link for more details, such as CVSS details, affected products, timeline, and more...

CVE ID :CVE-2026-48067 Published : June 22, 2026, 9:46 p.m. | 9 hours, 57 minutes ago Description :Filament is a collection of full-stack components for accelerated Laravel development. From filament/actions 4.0.0 until 4.11.4 and 5.6.4 and from filament/tables 3.0.0 until 3.3.51, the recordSelectOptionsQuery() method may be used to scope the options available in the Select field for AttachAction and AssociateAction. However, the built-in validation rule for these fields did not apply the same scope. As a result, a user who can trigger these actions could tamper with the Livewire component's state and submit an out-of-scope value. This vulnerability is fixed in filament/actions 4.11.4 and 5.6.4 and filament/tables 3.3.51. Severity: 6.5 | MEDIUM Visit the link for more details, such as CVSS details, affected products, timeline, and more...