CVE-2025-59536
📋 TL;DR
Claude Code versions before 1.0.111 contain a code injection vulnerability that allows arbitrary code execution when users start the application in untrusted directories. The vulnerability bypasses the startup trust dialog, enabling malicious code to run before user consent. Users with manual updates are affected, while auto-update users have received the fix.
💻 Affected Systems
- Claude Code
📦 What is this software?
Claude Code by Anthropic
⚠️ Risk & Real-World Impact
Worst Case
Complete system compromise via arbitrary code execution, potentially leading to data theft, ransomware deployment, or persistent backdoor installation.
Likely Case
Local privilege escalation or execution of malicious scripts from untrusted project directories, compromising user data and system integrity.
If Mitigated
No impact if users only run Claude Code in trusted directories or have updated to patched version.
🎯 Exploit Status
Exploitation requires social engineering to convince users to run Claude Code in untrusted locations.
🛠️ Fix & Mitigation
✅ Official Fix
Patch Version: 1.0.111
Vendor Advisory: https://github.com/anthropics/claude-code/security/advisories/GHSA-4fgq-fpq9-mr3g
Restart Required: Yes
Instructions:
1. Check current version with 'claude-code --version'. 2. If version <1.0.111, download latest from official source. 3. Install update. 4. Restart Claude Code.
🔧 Temporary Workarounds
Restrict directory usage
allOnly run Claude Code in trusted, controlled directories
Disable auto-start features
allPrevent Claude Code from automatically opening recent projects
🧯 If You Can't Patch
- Implement application whitelisting to prevent unauthorized code execution
- Restrict user permissions and implement least privilege access controls
🔍 How to Verify
Check if Vulnerable:
Check if Claude Code version is below 1.0.111 and if application runs code from untrusted directories without trust prompt.Check Version:
claude-code --versionVerify Fix Applied:
Update to version 1.0.111 and verify trust dialog appears before code execution in untrusted directories.📡 Detection & Monitoring
Log Indicators:
- Unexpected process execution from Claude Code
- Trust dialog bypass events
- Code execution from untrusted paths
Network Indicators:
- Unusual outbound connections from Claude Code process
SIEM Query:
Process creation where parent process is Claude Code and command line contains suspicious code execution patterns