Where
-Infinity
0

Trending

Last week
Last month
Last year

Vendor Risk Score

See how anthropic compares to other vendors in security performance

View Risk Score →

Escaping Claude Cowork’s local VM sandbox via CVE-2026-46331

First published (updated )
Social
reddit

BleepingComputerClaude Chrome extension flaw lets malicious extensions trigger AI actions

Exploited
First published (updated )
News
BleepingComputer

Dark ReadingClaude Flaw Automatically Sends Malicious Prompts to AI Agents

Exploited
First published (updated )
News
Dark Reading

Claude CodeClaude Code: Sandbox Escape via Git Worktree Path Confusion Allows Unsandboxed Code Execution

Risk 77
Severity
7.7
First published (updated )
CVE-2026-55607

CVE-2026-52870: Anthropic MCP Python SDK Missing Authorization Flaw

First published (updated )
Social
reddit
Free Weekly Intel

Don't miss critical vulnerabilities

Join thousands of security professionals who receive our weekly digest of trending CVEs, zero-days, and exploited vulnerabilities.

No spam. Unsubscribe anytime.

npm/@anthropic-ai/claude-codeClaude Code: Insecure Temporary File in /copy Command Enables Response Disclosure and Symlink-Based File Write

Risk 46
Severity
4.4
First published (updated )
CVE-2026-46406

Anthropic Claude Desktop CoworkAnthropic Claude Desktop Cowork VM Image Contents Not Validated Before Use

Risk 68
Severity
8.7
First published (updated )
CVE-2026-7574

npm/@anthropic-ai/claude-codeClaude Code: Out-of-Band Data Exfiltration via Pre-Approved HuggingFace Domain in WebFetch

Risk 66
Severity
6
First published (updated )
CVE-2026-54316

BleepingComputerWindows 11 and Microsoft Edge hacked at Pwn2Own Berlin 2026

Exploited
Zeroday
First published (updated )
News
BleepingComputer

Anthropic Claude DesktopClaude Desktop: Local Privilege Escalation via Directory Junction in CoworkVMService

Risk 73
Severity
8.5
First published (updated )
CVE-2026-44470
Free Weekly Intel

Don't miss critical vulnerabilities

Join thousands of security professionals who receive our weekly digest of trending CVEs, zero-days, and exploited vulnerabilities.

No spam. Unsubscribe anytime.

Anthropic Claude DesktopClaude Desktop: SSH Host Key Verification Bypass Allows Man-in-the-Middle Attack on Remote Sessions

Risk 52
Severity
7.4
First published (updated )
CVE-2026-44467

npm/@anthropic-ai/sdkClaude SDK for TypeScript has Insecure Default File Permissions in Local Filesystem Memory Tool

Risk 31
Severity
4.8
First published (updated )
CVE-2026-41686

npm/@anthropic-ai/claude-codeClaude Code arbitrary code execution via git worktree commondir trust dialog bypass

Risk 77
Severity
7.7
First published (updated )
CVE-2026-40068

Anthropic Claude CodeClaude Code: Sandbox Escape via Symlink Following Allows Arbitrary File Write Outside Workspace

Risk 87
Severity
7.7
First published (updated )
CVE-2026-39861

Claude CodeClaude Code: Insecure System-Wide Configuration Loading Enables Local Privilege Escalation on Windows

Risk 64
Severity
5.4
First published (updated )
CVE-2026-35603
Free Weekly Intel

Don't miss critical vulnerabilities

Join thousands of security professionals who receive our weekly digest of trending CVEs, zero-days, and exploited vulnerabilities.

No spam. Unsubscribe anytime.

The RegisterMCP 'design flaw' puts 200k servers at risk: Researcher

Exploited
First published (updated )
News
The Register

Anthropic ClaudeAnthropic Claude Code & Agent SDK OS Command Injection via Authentication Helper

Risk 44
Severity
9.3
First published (updated )
CVE-2026-35022

Anthropic Claude Code CLIAnthropic Claude Code & Agent SDK OS Command Injection via TERMINAL Environment Variable

Risk 41
Severity
8.6
First published (updated )
CVE-2026-35020

npm/claude-sdkClaude SDK for TypeScript: Memory Tool Path Validation Allows Sandbox Escape to Sibling Directories

Risk 37
Severity
6.3
First published (updated )
CVE-2026-34451

pypi/anthropicClaude SDK for Python: Insecure Default File Permissions in Local Filesystem Memory Tool

Risk 31
Severity
4.8
First published (updated )
CVE-2026-34450
Free Weekly Intel

Don't miss critical vulnerabilities

Join thousands of security professionals who receive our weekly digest of trending CVEs, zero-days, and exploited vulnerabilities.

No spam. Unsubscribe anytime.

pypi/anthropicClaude SDK for Python: Memory Tool Path Validation Race Condition Allows Sandbox Escape

Risk 39
Severity
5.8
First published (updated )
CVE-2026-34452

Anthropic Claude for Windows (Claude Setup.exe installer)Uncontrolled search path elements in Anthropic Claude for Windows installer (Claude Setup.exe) versi…

Risk 68
Severity
4.7
First published (updated )
CVE-2026-22561

Anthropic's Claude Code CLI had a workspace trust bypass (CVE-2026-33068). Repository settings loaded before trust dialog. Classic configuration loading order bug in an AI developer tool

First published (updated )
Social
reddit

Claude Code workspace trust dialog bypass via repository settings loading order [CVE-2026-33068, CVSS 7.7]. Settings resolved before trust dialog shown.

First published (updated )
Social
reddit

npm/@anthropic-ai/claude-codeClaude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File

Risk 58
Severity
7.7
EPSS
0.14%
First published (updated )
CVE-2026-33068
Free Weekly Intel

Don't miss critical vulnerabilities

Join thousands of security professionals who receive our weekly digest of trending CVEs, zero-days, and exploited vulnerabilities.

No spam. Unsubscribe anytime.

The RegisterMicrosoft Azure CTO set Claude on his 1986 Apple II code, says it found vulns

Exploited
First published (updated )
News
The Register

The RegisterClaude's collaboration tools allowed remote code execution

Exploited
First published (updated )
News
The Register

I went through every AI agent security incident from 2025 and fact-checked all of it. Here is what was real, what was exaggerated, and what the CrewAI and LangGraph docs will never tell you.

First published (updated )
Social
reddit

ZDNetHow Microsoft obliterated safety guardrails on popular AI models - with just one prompt

First published (updated )
News
ZDNet

Claude Claude CodeClaude Code Has Sandbox Escape via Persistent Configuration Injection in settings.json

Risk 63
Severity
7.7
EPSS
0.06%
First published (updated )
CVE-2026-25725
Free Weekly Intel

Don't miss critical vulnerabilities

Join thousands of security professionals who receive our weekly digest of trending CVEs, zero-days, and exploited vulnerabilities.

No spam. Unsubscribe anytime.

Company

Resources

Contact

SecAlerts Pty Ltd.
132 Wickham Terrace
Fortitude Valley,
QLD 4006, Australia
info@secalerts.co
By using SecAlerts services, you agree to our services end-user license agreement. This website is safeguarded by reCAPTCHA and governed by the Google Privacy Policy and Terms of Service. All names, logos, and brands of products are owned by their respective owners, and any usage of these names, logos, and brands for identification purposes only does not imply endorsement. If you possess any content that requires removal, please get in touch with us.
© 2026 SecAlerts Pty Ltd.
ABN: 70 645 966 203, ACN: 645 966 203