Lucene search
K

30 matches found

Github Security Blog
Github Security Blog
added 2026/04/24 7:30 p.m.31 views

Gemini CLI: Remote Code Execution via workspace trust and tool allowlisting bypasses

Summary Gemini CLI @google/gemini-cli and the run-gemini-cli GitHub Action are being updated to harden workspace trust and tool allowlisting, in particular when used in untrusted environments like GitHub Actions. This update introduces a breaking change to how non-interactive headless environment...

6.5AI score
Exploits0References2Affected Software2
Cvelist
Cvelist
added 2026/03/20 8:17 a.m.22 views

CVE-2026-33068 Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File

Claude Code is an agentic coding tool. Versions prior to 2.1.53 resolved the permission mode from settings files, including the repo-controlled .claude/settings.json, before determining whether to display the workspace trust confirmation dialog. A malicious repository could set...

7.7CVSS0.00235EPSS
Exploits0References1
CVE
CVE
added 2026/03/20 8:17 a.m.415 views

CVE-2026-33068

CVE-2026-33068 affects Claude Code. Versions prior to 2.1.53 could bypass the workspace trust dialog by using permissions.defaultMode set to bypassPermissions in the repo-controlled .claude/settings.json, allowing silent trust mode on first open and enabling tool execution without user consent. T...

8.8CVSS5.9AI score0.00235EPSS
Exploits0References1Affected Software1
OSV
OSV
added 2026/03/20 8:17 a.m.3 views

CVE-2026-33068 Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File

Claude Code is an agentic coding tool. Versions prior to 2.1.53 resolved the permission mode from settings files, including the repo-controlled .claude/settings.json, before determining whether to display the workspace trust confirmation dialog. A malicious repository could set...

7.7CVSS6AI score0.00235EPSS
Exploits0References3
Vulnrichment
Vulnrichment
added 2026/03/20 8:17 a.m.0 views

CVE-2026-33068 Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File

Claude Code is an agentic coding tool. Versions prior to 2.1.53 resolved the permission mode from settings files, including the repo-controlled .claude/settings.json, before determining whether to display the workspace trust confirmation dialog. A malicious repository could set...

7.7CVSS5.9AI score0.00235EPSS
Exploits0References1
CNNVD
CNNVD
added 2026/03/20 12:0 a.m.3 views

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 2.1.53 contained a security vulnerability, which was caused by improper order of permission mode parsing. This vulnerability could potentially allow bypassing the workspace...

8.8CVSS5.8AI score0.00235EPSS
Exploits0References1
Github Security Blog
Github Security Blog
added 2026/03/19 12:42 p.m.8 views

Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File

Claude Code resolved the permission mode from settings files, including the repo-controlled .claude/settings.json, before determining whether to display the workspace trust confirmation dialog. A malicious repository could set permissions.defaultMode to bypassPermissions in its committed...

8.8CVSS5.9AI score0.00235EPSS
Exploits0References3Affected Software1
Snyk
Snyk
added 2026/03/19 12:42 p.m.1 views

Reliance on Untrusted Inputs in a Security Decision

Overview @anthropic-ai/claude-code is an Use Claude, Anthropic's AI assistant, right from your terminal. Claude can understand your codebase, edit files, run terminal commands, and handle entire workflows for you. Affected versions of this package are vulnerable to Reliance on Untrusted Inputs in...

8.8CVSS6AI score0.00235EPSS
Exploits0References2
OSV
OSV
added 2026/03/19 12:42 p.m.4 views

GHSA-MMGP-WC2J-QCV7 Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File

Claude Code resolved the permission mode from settings files, including the repo-controlled .claude/settings.json, before determining whether to display the workspace trust confirmation dialog. A malicious repository could set permissions.defaultMode to bypassPermissions in its committed...

7.7CVSS5.9AI score0.00235EPSS
Exploits0References3
EUVD
EUVD
added 2026/03/17 9:31 p.m.1 views

EUVD-2026-12638

Improper trust boundary enforcement in Kiro IDE before version 0.8.0 on all supported platforms might allow a remote unauthenticated threat actor to execute arbitrary code via maliciously crafted project directory files that bypass workspace trust protections when a local user opens the directory...

8.5CVSS6.2AI score0.00029EPSS
Exploits0References3
CVE
CVE
added 2026/03/17 7:11 p.m.5 views

CVE-2026-4295

CVE-2026-4295 affects Kiro IDE prior to 0.8.0. Improper trust boundary enforcement may allow a remote unauthenticated actor to execute arbitrary code via maliciously crafted project directory files that bypass workspace trust protections when a local user opens the directory. Affected software: K...

8.5CVSS6.2AI score0.00029EPSS
Exploits0References2
Vulnrichment
Vulnrichment
added 2026/03/17 7:11 p.m.1 views

CVE-2026-4295 Arbitrary code execution via crafted project files in Kiro IDE

Improper trust boundary enforcement in Kiro IDE before version 0.8.0 on all supported platforms might allow a remote unauthenticated threat actor to execute arbitrary code via maliciously crafted project directory files that bypass workspace trust protections when a local user opens the directory...

8.5CVSS6.2AI score0.00029EPSS
Exploits0References2
ATTACKERKB
ATTACKERKB
added 2026/03/17 7:11 p.m.2 views

CVE-2026-4295

Improper trust boundary enforcement in Kiro IDE before version 0.8.0 on all supported platforms might allow a remote unauthenticated threat actor to execute arbitrary code via maliciously crafted project directory files that bypass workspace trust protections when a local user opens the directory...

8.5CVSS6.2AI score0.00029EPSS
Exploits0References3Affected Software1
NVD
NVD
added 2026/02/09 11:16 p.m.4 views

CVE-2026-25931

vscode-spell-checker is a basic spell checker that works well with code and documents. Prior to v4.5.4, DocumentSettings.determineIsTrusted treats the configuration value cSpell.trustedWorkspace as the authoritative trust flag. The value defaults to true package.json and is read from workspace...

7.8CVSS0.00021EPSS
Exploits0References4
Vulnrichment
Vulnrichment
added 2026/02/09 10:10 p.m.3 views

CVE-2026-25931 vscode-spell-checker has a workspace-trust bypass Code Execution

vscode-spell-checker is a basic spell checker that works well with code and documents. Prior to v4.5.4, DocumentSettings.determineIsTrusted treats the configuration value cSpell.trustedWorkspace as the authoritative trust flag. The value defaults to true package.json and is read from workspace...

7.8CVSS5.7AI score0.00021EPSS
Exploits0References4
CVE
CVE
added 2026/02/09 10:10 p.m.14 views

CVE-2026-25931

The vulnerability affects the vscode-spell-checker extension prior to version 4.5.4. It arises because DocumentSettings._determineIsTrusted uses the cSpell.trustedWorkspace setting as the authoritative trust flag, defaulting to true in package.json. This allows an untrusted workspace to cause the...

7.8CVSS5.7AI score0.00021EPSS
Exploits0References4
Cvelist
Cvelist
added 2026/02/09 10:10 p.m.23 views

CVE-2026-25931 vscode-spell-checker has a workspace-trust bypass Code Execution

vscode-spell-checker is a basic spell checker that works well with code and documents. Prior to v4.5.4, DocumentSettings.determineIsTrusted treats the configuration value cSpell.trustedWorkspace as the authoritative trust flag. The value defaults to true package.json and is read from workspace...

7.8CVSS0.00021EPSS
Exploits0References4
EUVD
EUVD
added 2025/10/03 8:7 p.m.2 views

EUVD-2025-27563

Malicious code in bioql PyPI...

8.7CVSS6.5AI score0.00529EPSS
Exploits0References3
RedhatCVE
RedhatCVE
added 2025/09/12 3:28 p.m.3 views

CVE-2025-59041

Claude Code is an agentic coding tool. At startup, Claude Code executed a command templated in with git config user.email. Prior to version 1.0.105, a maliciously configured user email in git could be used to trigger arbitrary code execution before a user accepted the workspace trust dialog. User...

8.7CVSS8AI score0.00529EPSS
Exploits0References1
The Hacker News
The Hacker News
added 2025/09/12 4:49 a.m.15 views

Cursor AI Code Editor Flaw Enables Silent Code Execution via Malicious Repositories

A security weakness has been disclosed in the artificial intelligence AI-powered code editor Cursor that could trigger code execution when a maliciously crafted repository is opened using the program. The issue stems from the fact that an out-of-the-box security setting is disabled by default,...

9.3CVSS8.3AI score0.00263EPSS
Exploits3
Rows per page
Query Builder