Sonar

Mentions · 7d0
Mentions · 30d0
All mentions1
First seenSep 2, 2026
Last mentionedSep 2, 2026

Timeline

1 articles · newest first

Malicious .git Configs Can Make Claude, Codex, Cursor, and Other AI Agents Run Attacker Code

Malicious .git core.fsmonitor configs let attacker code run in AI coding agents like Claude Code and Codex; four of eight flaws remain unpatched.

The Hacker NewsSep 2, 14:06 UTC · Sep 2, 2026VulnerabilityCVE-2026-19592 CVE-2026-72718 CVE-2026-71963+37 · 3 reads

Related CVEs

by mentions

CVE-2026-71963<1%8.6
Command injection RCE in Hermes Agent via malicious .git/config
published Sep 3, 2026Hermes AgentPoC
CVE-2022-24346<1%7.8
In JetBrains IntelliJ IDEA before 2021.3.1, local code execution via RLO (Right-to-Left Override) characters was possible
published Feb 25, 2022jetbrains intellij idea
CVE-2021-4389113%7.8
Visual Studio Code Remote Code Execution Vulnerability
published Dec 15, 2021microsoft visual studio code
CVE-2026-55607<1%7.7
Claude Code: Claude Code's worktree handling allowed creation of worktrees named ".git" and navigation to worktrees outside the sandbox…
published Jun 29, 2026anthropic claude code
CVE-2026-19592<1%7.3
Unsandboxed attacker code execution in OpenAI Codex via malicious Git core.fsmonitor
published Sep 1, 2026OpenAI Codex CLI (Windows, macOS, and Linux)
CVE-2026-72718<1%7.0
goose is general-purpose AI agent that runs on your machine
published Aug 10, 2026

Appears with

gooseproduct · 1
MITREorganization · 1
Visual Studio Codeproduct · 1
Nous Researchorganization · 1
VulnCheckvendor · 1
Anthropicvendor · 1
xAIorganization · 1
Codexproduct · 1
Gitproduct · 1
Manifold Securityvendor · 1
Alibabavendor · 1
Grok Buildproduct · 1
Claude Codeproduct · 1
OpenAIvendor · 1
Cursorproduct · 1
JetBrains IDEsproduct · 1

Entities are extracted by the model from each article. First seen Sep 2, 2026, last mentioned Sep 2, 2026. Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.