PULSE
FEED
ransompanzer reclama a Asesoría FAR · Professional Servicesransompanzer reclama a Ressources Si · FR · Professional Servicesransomstorm reclama a First Secure Bank Group · US · Financial Servicesransomqilin reclama a XICO · MX · Not Foundransomqilin reclama a Island · CA · Technologyransomqilin reclama a Revenga Smart Solutions · ES · Technologyransomqilin reclama a Willatt & Flickinger · US · Not Foundransomarcusmedia reclama a Pantaneiro Capas · BR · Manufacturingransomemperador reclama a Car Service Abschlepp · DE · TransportationvulnKEV agrega CVE-2026-88772 — Citrix / NetScalervulnKEV agrega CVE-2026-88771 — Citrix / NetScalerransomm3rx reclama a cipher.systems · US · Technologyransombarracuda reclama a International Chemical Co. · Manufacturingransompayoutsking reclama a M****n · US · Not Foundransompanzer reclama a Asesoría FAR · Professional Servicesransompanzer reclama a Ressources Si · FR · Professional Servicesransomstorm reclama a First Secure Bank Group · US · Financial Servicesransomqilin reclama a XICO · MX · Not Foundransomqilin reclama a Island · CA · Technologyransomqilin reclama a Revenga Smart Solutions · ES · Technologyransomqilin reclama a Willatt & Flickinger · US · Not Foundransomarcusmedia reclama a Pantaneiro Capas · BR · Manufacturingransomemperador reclama a Car Service Abschlepp · DE · TransportationvulnKEV agrega CVE-2026-88772 — Citrix / NetScalervulnKEV agrega CVE-2026-88771 — Citrix / NetScalerransomm3rx reclama a cipher.systems · US · Technologyransombarracuda reclama a International Chemical Co. · Manufacturingransompayoutsking reclama a M****n · US · Not Found
← Todos los CVEs
CVE Watch26 sept 2026

CVE-2026-100652

vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-

CVSS

5.9

Medio

EPSS

0.3%

p23

KEV

—

Exploit Today

7

0-100

Publicado: 26 sept 2026 · Última mod.: 26 sept 2026 · CWE-20

EPSS · 30d

Sin historial EPSS suficiente todavía.

Descripción técnica

vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-of-vocabulary token IDs to reach MinTokensLogitsProcessor. Attackers can submit requests with min_tokens greater than zero and out-of-vocabulary stop_token_ids to trigger CUDA tensor indexing failures that leave EngineCore in a fatal state requiring service restart.

Referencias oficiales
CVEs relacionados
CVECVSSEPSSKEVRExplotTítuloVis.
CVE-2026-88771—
—
KEV—50Citrix NetScaler Improper Input Validation Vulnerability6h
CVE-2026-101041—
—
——0The account recovery (password reset) functionality in the vulnerability-lookup web application contains a time-of-check-to-time-of-use (TOCTOU) race condition in the consumption of single-use recovery tokens. The original implementation verified the token nonce against the stored digest and then consumed (cleared) it in separate database operations. Two concurrent HTTP requests presenting the same valid recovery token could both pass the verification check before either transaction committed, allowing both to set their own password on the target account. The last transaction to commit overwrites the first, enabling an attacker who possesses a valid recovery token to replace the legitimate user's password with one of their choosing. A secondary defect in the same endpoint (confirm_account) allowed a valid recovery link to be used to set an empty or trivially short password (e.g., three characters). The view handler performed only a manual equality comparison between the two password fields and never invoked the form's validation logic, bypassing the intended minimum-length and complexity constraints. The affected component is the user account recovery endpoint (/user/confirm_account/<token>) and the associated token verification and consumption logic in the User model (website/models/user.py) and the view layer (website/web/views/user.py).14h
CVE-2026-1008338.2 ALT
12.0%
——4Contrast (edgelesssys/contrast) versions 1.14.0 before 1.23.1 generate runtime policies that fail to detect all container image substitutions. A bad rebase during a Kata Containers update accidentally introduced an `allow_storage` rule that accepts storage entries using the `image_guest_pull` driver without verifying the image digest. An attacker with access to the Kata agent API — for example, a Kubernetes cluster administrator in Contrast's threat model — can therefore substitute a container image with an exploit payload, provided the substituted image satisfies the remaining policy rules, undermining the confidential container's integrity guarantees.1d
CVE-2026-1006995.3 MED
8.0%
——2Nodemailer is a Node.js email-sending library. In versions >= 9.1.0 and < 10.0.9, the address parser (src/addressparser) mishandles addresses whose local-part is a quoted string and that are followed by RFC 5322 comments, allowing trailing comment-separated domain atoms to be retained in the normalized address. For example, the input "user"@example.com(x)evil.com is parsed to the address value 'user@example.com evil.com', which contains additional attacker-controlled domain text separated by a literal space. This parsed value is used without further strict recipient validation when the message envelope is built (envelope.to in src/mime-node), so a malformed/ambiguous recipient address can be accepted and placed in the SMTP envelope. Whether this results in delivery to an unintended recipient on real SMTP servers has not been confirmed. The issue is a variant of the RFC 5322 comment parsing problem addressed in GHSA-cc9r-2j5m-2m83, affecting the separate quoted-local-part code path. Version 10.0.9 contains a fix.2d
CVE-2026-1006647.5 ALT
23.5%
——7Netty's HTTP/3 codec (io.netty:netty-codec-http3) versions 4.2.2.Final through 4.2.17.Final builds the HTTP/3 :authority pseudo-header from the HTTP/1 Host header before considering the authority of an absolute-form HTTP/1 request-target. In HttpConversionUtil.toHttp3Headers(HttpMessage, boolean) — reached via Http3FrameToHttpObjectCodec(false) — a non-empty Host header takes precedence over the request-target authority, contrary to the HTTP/1.1 rule that a server receiving an absolute-form request-target must ignore the Host header. In a Netty-based HTTP/1-to-HTTP/3 gateway, proxy, or protocol bridge, a remote client can send a request such as "GET https://trusted.example/admin HTTP/1.1" with "Host: attacker.example", causing components that validate, authorize, or route on the RFC-defined request-target authority to reach a different decision than the upstream HTTP/3 peer, which receives :authority derived from the conflicting Host header. This authority confusion can affect virtual-host routing, allow-list checks, backend selection, cache keys, and URL generation. The advisory reports integrity impact only (no code execution, memory corruption, or availability impact). Fixed in 4.2.18.Final.12h
CVE-2026-1006637.5 ALT
20.5%
——6Netty's HTTP/3 codec (io.netty:netty-codec-http3) from 4.2.2.Final through 4.2.17.Final does not special-case HTTP/1 CONNECT authority-form request-targets when converting HTTP/1 messages to HTTP/3 in HttpConversionUtil.toHttp3Headers. The authority-form target (e.g., "CONNECT trusted.example:443") is parsed as a URI, so its host is emitted as :scheme, :path is set to "/", and the HTTP/1 Host header is used as :authority; if no Host header is present the CONNECT target is dropped. In a Netty-based HTTP/1-to-HTTP/3 proxy or gateway, a remote client can send a CONNECT request whose Host header names a different authority than the request-target, producing a malformed HTTP/3 CONNECT whose tunnel :authority is attacker-controlled. This can bypass tunnel allow-lists, egress policy, backend selection, or audit controls that validate the HTTP/1 CONNECT request-target before forwarding over HTTP/3. The issue is fixed in 4.2.18.Final.12h