Evaluating and Enhancing the Vulnerability Reasoning Capabilities of Large Language Models

Large Language Models LLMs have demonstrated remarkable proficiency in vulnerability detection. However, a critical reliability gap persists: models frequently yield correct detection verdicts based on hallucinated logic or superficial patterns that deviate from the actual root cause. This...

Claude Code 操作系统命令注入漏洞

Claude Code is an open-source terminal-native AI programming tool developed by Anthropic. Versions of Claude Code prior to 2.0.55 contained a vulnerability related to operating system command injection. This vulnerability stemmed from insufficient validation of commands that utilized the echo...

Claude Code 安全漏洞

Claude Code is an open-source terminal-native AI programming tool developed by Anthropic. Versions of Claude Code prior to 2.1.2 contained a security vulnerability. This vulnerability stemmed from the bubblewrap sandbox mechanism, which failed to properly protect the.claude/settings.json file whe...

PT-2026-6766

Name of the Vulnerable Software and Affected Versions Claude Code versions prior to 2.1.2 Description Claude Code, an agentic coding tool, had a flaw in its bubblewrap sandboxing mechanism. The mechanism did not adequately protect the .claude/settings.json configuration file when it was absent at...

CVE-2026-25722: Improper Input Validation

Claude Code is an agentic coding tool. Prior to version 2.0.57, Claude Code failed to properly validate directory changes when combined with write operations to protected folders. By using the cd command to navigate into sensitive directories like .claude, it was possible to bypass write protecti...

PT-2026-6862

Claude Code failed to properly validate commands using piped sed operations with the echo command, allowing attackers to bypass file write restrictions. This vulnerability enabled writing to sensitive directories like the .claude folder and paths outside the project scope. Exploiting this require...

CVE-2026-25723: Improper Input Validation

Claude Code is an agentic coding tool. Prior to version 2.0.55, Claude Code failed to properly validate commands using piped sed operations with the echo command, allowing attackers to bypass file write restrictions. This vulnerability enabled writing to sensitive directories like the .claude...

CVE-2026-25725: Trust Boundary Violation

Claude Code is an agentic coding tool. Prior to version 2.1.2, Claude Code's bubblewrap sandboxing mechanism failed to properly protect the .claude/settings.json configuration file when it did not exist at startup. While the parent directory was mounted as writable and .claude/settings.local.json...

CVE-2026-25724: UNIX Symbolic Link (Symlink) Following

Claude Code is an agentic coding tool. Prior to version 2.1.7, Claude Code failed to strictly enforce deny rules configured in settings.json when accessing files through symbolic links. If a user explicitly denied Claude Code access to a file such as /etc/passwd and Claude Code had access to a...

PT-2026-6764

Name of the Vulnerable Software and Affected Versions Claude Code versions prior to 2.0.55 Description Claude Code, an agentic coding tool, exhibited a flaw in command validation. Specifically, the software did not adequately validate commands utilizing piped sed operations with the echo command...

Claude Code 操作系统命令注入漏洞

Claude Code is an open-source terminal-native AI programming tool developed by Anthropic. Versions of Claude Code prior to 2.0.57 contained a vulnerability related to operating system command injection. This vulnerability arose from an inability to properly validate directory changes in conjuncti...

PT-2026-6763

Name of the Vulnerable Software and Affected Versions Claude Code versions prior to 2.0.57 Description Claude Code, an agentic coding tool, did not properly validate directory changes when combined with write operations to protected folders. Utilizing the cd command to navigate into sensitive...

Claude Code 授权问题漏洞

Claude Code is an open-source terminal-native AI programming tool developed by Anthropic. Versions of Claude Code prior to 2.1.7 had an authorization issue vulnerability. This vulnerability occurred due to the failure to strictly enforce the denial rules configured in settings.json when accessing...

PT-2026-6765

Name of the Vulnerable Software and Affected Versions Claude Code versions prior to 2.1.7 Description Claude Code, an agentic coding tool, did not properly enforce deny rules defined in the settings.json file when handling symbolic links. Specifically, if access to a file like /etc/passwd was...

CVE-2026-24887

Claude Code is an agentic coding tool. Prior to version 2.0.72, due to an error in command parsing, it was possible to bypass the Claude Code confirmation prompt to trigger execution of untrusted commands through the find command. Reliably exploiting this required the ability to add untrusted...

CVE-2026-24052

Claude Code is an agentic coding tool. Prior to version 1.0.111, Claude Code contained insufficient URL validation in its trusted domain verification mechanism for WebFetch requests. The application used a startsWith function to validate trusted domains e.g., docs.python.org,...

CVE-2026-24053

Claude Code is an agentic coding tool. Prior to version 2.0.74, due to a Bash command validation flaw in parsing ZSH clobber syntax, it was possible to bypass directory restrictions and write files outside the current working directory without user permission prompts. Exploiting this required the...

CVE-2026-24052

Claude Code is an agentic coding tool. Prior to version 1.0.111, Claude Code contained insufficient URL validation in its trusted domain verification mechanism for WebFetch requests. The application used a startsWith function to validate trusted domains e.g., docs.python.org,...

CVE-2026-24053

Claude Code is an agentic coding tool. Prior to version 2.0.74, due to a Bash command validation flaw in parsing ZSH clobber syntax, it was possible to bypass directory restrictions and write files outside the current working directory without user permission prompts. Exploiting this required the...

CVE-2026-24887

Claude Code is an agentic coding tool. Prior to version 2.0.72, due to an error in command parsing, it was possible to bypass the Claude Code confirmation prompt to trigger execution of untrusted commands through the find command. Reliably exploiting this required the ability to add untrusted...