PT-2026-81848

Name of the Vulnerable Software and Affected Versions dbgpt-app versions prior to 0.8.1 Description An issue exists in the skill upload endpoint within packages/dbgpt-app/src/dbgpt app/openapi/api v1/agentic data api.py where the destination path for uploaded skills is constructed using the...

CVE-2026-73034

DB-GPT v0.8.1 contains an unauthenticated path traversal vulnerability that allows remote attackers to write arbitrary files to any location on the server by injecting directory traversal sequences into the userid HTTP header of the Python file-upload endpoint. Attackers can send a crafted...

CVE-2026-73034

DB-GPT v0.8.1 is affected by an unauthenticated path traversal flaw in the Python file-upload endpoint. By manipulating the user_id HTTP header with directory traversal sequences, an attacker can write arbitrary files to locations outside the intended upload directory (e.g., Python startup hooks,...

CVE-2026-73034 DB-GPT v0.8.1 Path Traversal Arbitrary File Write via user_id Header

DB-GPT v0.8.1 contains an unauthenticated path traversal vulnerability that allows remote attackers to write arbitrary files to any location on the server by injecting directory traversal sequences into the userid HTTP header of the Python file-upload endpoint. Attackers can send a crafted...

CVE-2026-73034

DB-GPT v0.8.1 contains an unauthenticated path traversal vulnerability that allows remote attackers to write arbitrary files to any location on the server by injecting directory traversal sequences into the userid HTTP header of the Python file-upload endpoint. Attackers can send a crafted...

CVE-2026-73034 DB-GPT v0.8.1 Path Traversal Arbitrary File Write via user_id Header

DB-GPT v0.8.1 contains an unauthenticated path traversal vulnerability that allows remote attackers to write arbitrary files to any location on the server by injecting directory traversal sequences into the userid HTTP header of the Python file-upload endpoint. Attackers can send a crafted...

EUVD-2026-56925

DB-GPT v0.8.1 contains an unauthenticated path traversal vulnerability that allows remote attackers to write arbitrary files to any location on the server by injecting directory traversal sequences into the userid HTTP header of the Python file-upload endpoint. Attackers can send a crafted...

PT-2026-70890

Name of the Vulnerable Software and Affected Versions DB-GPT version 0.8.1 Description An unauthenticated path traversal issue exists in the Python file-upload endpoint. Remote attackers can write arbitrary files to any location on the server by injecting directory traversal sequences into the us...

Linux Distros Unpatched Vulnerability : CVE-2026-20345

The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - 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...

CVE-2026-20345

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...

DEBIAN-CVE-2026-20345

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...

AZL-95444 CVE-2026-20345 affecting package clamav 1.5.3-1

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...

UBUNTU-CVE-2026-20345

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...

CVE-2026-20345

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...

EUVD-2026-54507

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...

CVE-2026-20345

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...

CVE-2026-20345

ClamAV's GPT file format parser is affected by memory corruption due to improper endian conversion, causing an out-of-bounds write. An unauthenticated attacker can submit a crafted GPT file to a vulnerable ClamAV scanner, potentially terminating the scanning process and causing a DoS. CVSSv3.1 ba...

EUVD-2026-54301

A vulnerability was identified in FoundationAgents MetaGPT up to 0.8.2. This impacts an unknown function. Such manipulation leads to code injection. The attack needs to be performed locally. The exploit is publicly available and might be used. The vendor was contacted early about this disclosure...

PT-2026-69073

Name of the Vulnerable Software and Affected Versions ClamAV affected versions not specified Description A flaw in the GPT file format parser allows an unauthenticated remote attacker to cause a Denial of Service DoS or other impacts via memory corruption. The issue stems from improper handling o...

Anthropic’s Opus 5 Is Better at Resisting Prompt Injection

The chart is interesting. On the IPI benchmark, Opus 5 improved over Opus 4.8, reducing the probability of an attacker succeeding within 15 attempts from 5.5% to 2.0%, and from 0.5% to 0.2% on 1 attempt. It also improved on Sonnet 5 5.9% at k=15 and Mythos 5 2.6%, making it the most robust model...