CVE-2026-56787

RTKLIB through 2.4.3 contains an off-by-one out-of-bounds read vulnerability in the decodessr3 function at src/rtcm3.c:1446 that allows remote attackers to trigger a global buffer overflow via crafted RTCM3 SSR messages with attacker-controlled signal mode fields. Remote attackers can exploit thi...

EUVD-2026-39529

RTKLIB through 2.4.3 contains an off-by-one out-of-bounds read vulnerability in the decodessr3 function at src/rtcm3.c:1446 that allows remote attackers to trigger a global buffer overflow via crafted RTCM3 SSR messages with attacker-controlled signal mode fields. Remote attackers can exploit thi...

CVE-2026-56786 RTKLIB 2.4.3 - Out-of-bounds Write in decode_type1033 via Crafted RTCM3 Message

RTKLIB through 2.4.3 contains an out-of-bounds write vulnerability in decodetype1033 function that fails to clamp length counters to destination buffer size, allowing up to 191-byte overflow into fixed 64-byte descriptor fields. An attacker controlling an NTRIP or serial RTCM3 correction stream c...

CVE-2026-56770 libais 0.15 - Out-of-bounds Vector Access in VdmStream::AddLine via Invalid Sequential Message ID

libais through 0.15 VdmStream::AddLine uses an unchecked sentinel value as a vector index when processing AIS sentences with empty or out-of-range sequential message IDs. Remote attackers can crash services or vessel systems by sending crafted AIVDM sentences over VHF marine radio or IP feeds,...

CVE-2026-55967

CVE-2026-55967 covers AES-GCM streaming APIs that fail to reject extremely large cumulative single messages (>64 GiB), allowing counter wrap and keystream reuse and enabling plaintext recovery. Public documents reference the same issue across multiple OS advisories (Ubuntu, Debian, Debian-deri...

CVE-2026-55967

AES-GCM encryption/decryption with extremely large cumulative single message sizes 64 GiB were not properly rejected by the streaming APIs, allowing counter wrap, keystream reuse, and consequent plaintext recovery...

net/textproto: golang: Golang net/textproto: Misleading error messages via input injection

A flaw was found in the net/textproto package in Golang. When functions in this package return errors, they include their input as part of the error message. An attacker could exploit this by injecting misleading content into these error messages, which are then printed or logged. This could lead...

CVE-2026-54029

LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.4-rc1, the DELETE /api/messages/:conversationId/:messageId endpoint allows any authenticated user to delete any other user's messages. The validateMessageReq middleware only validates that the conversationId...

CVE-2026-4526

In EmberZNet v9.0.2 and earlier, malformed global ZCL messages can trigger out-of-bounds reads in framework parsing logic and terminate the process. These messages must come from a device that has already joined the network, and no information leakage back to the sender was observed...

CVE-2026-47151

In EmberZNet v9.0.2 and earlier, malformed ClearWeekdaySchedule messages can trigger out-of-bounds writes into Door Lock schedule state. The size and location of this data is limited. These messages must come from a device that has already joined the network. Only devices supporting the Door Lock...

EUVD-2026-39406

In EmberZNet v9.0.2 and earlier, malformed ClearWeekdaySchedule messages can trigger out-of-bounds writes into Door Lock schedule state. The size and location of this data is limited. These messages must come from a device that has already joined the network. Only devices supporting the Door Lock...

CVE-2026-47150

In EmberZNet v9.0.2 and earlier, malformed IAS Zone enrollment messages can trigger an out-of-bounds state-table write and terminate the process. The size and location of this write is limited. These messages must come from a device that has already joined the network. Only devices supporting the...

CVE-2026-47146

CVE-2026-47146 affects EmberZNet v9.0.2 and earlier; malformed Color Control messages can trigger asserts that abort the process. Impact is limited to devices that have already joined the network and that support the Color Control cluster. The provided documents do not specify a patch version or ...

CVE-2026-47145

In EmberZNet v9.0.2 and earlier, malformed Color Control messages can trigger asserts that terminate the process. The issue affects devices that already joined the network and that support the Color Control cluster. The problem is caused by malformed Color Control messages and results in an appli...

EUVD-2026-39399

In EmberZNet v9.0.2 and earlier, malformed Color Control messages can lead to asserts that terminate the process. These messages must come from a device that has already joined the network. Only devices supporting the Color Control cluster may be impacted...

CVE-2026-47145 Color Control hue/saturation assertion abort in EmberZNet v9.0.2

In EmberZNet v9.0.2 and earlier, malformed Color Control messages can lead to asserts that terminate the process. These messages must come from a device that has already joined the network. Only devices supporting the Color Control cluster may be impacted...

CVE-2026-47145

In EmberZNet v9.0.2 and earlier, malformed Color Control messages can lead to asserts that terminate the process. These messages must come from a device that has already joined the network. Only devices supporting the Color Control cluster may be impacted...

CVE-2026-4526

EmberZNet v9.0.2 and earlier has a vulnerability in the global ZCL command parser due to missing minimum-length validation, which can cause out-of-bounds reads in the framework parsing logic and terminate the process. The issue requires messages to originate from a device that has already joined ...

CVE-2026-4526

In EmberZNet v9.0.2 and earlier, malformed global ZCL messages can trigger out-of-bounds reads in framework parsing logic and terminate the process. These messages must come from a device that has already joined the network, and no information leakage back to the sender was observed...

CVE-2026-4526 Global ZCL command parser missing minimum-length validation in EmberZNet v9.0.2

In EmberZNet v9.0.2 and earlier, malformed global ZCL messages can trigger out-of-bounds reads in framework parsing logic and terminate the process. These messages must come from a device that has already joined the network, and no information leakage back to the sender was observed...