CVE-2026-21852

Claude Code Leaks Data via Malicious Environment Configuration Before Trust Confirmation

CVSS 5.3 medium · anthropics claude-code · published 2026-01-21

Claude Code is an agentic coding tool. Prior to version 2.0.65, vulnerability in Claude Code's project-load flow allowed malicious repositories to exfiltrate data including Anthropic API keys before users confirmed trust. An attacker-controlled repository could include a settings file that sets ANTHROPIC_BASE_URL to an attacker-controlled endpoint and when the repository was opened, Claude Code would read the configuration and immediately issue API requests before showing the trust prompt, potentially leaking the user's API keys. Users on standard Claude Code auto-update have received this fix already. Users performing manual updates are advised to update to version 2.0.65, which contains a patch, or to the latest version.

NVD · CVE.org

Coverage 1 source

1. 2026-02-25 Check Point Research critical Caught in the Hook: RCE and API Token Exfiltration Through Claude Code Project Files | CVE-2025-59536 | CVE-2026-21852

   Critical vulnerabilities in Anthropic's Claude Code allow RCE and API token exfiltration.

Mentioned with

CVE-2025-59536

Also covered

• CVE-2026-86950 Apple iOS and iPadOS CVSS 8.8 · 1 article · 2026-09-29
• CVE-2023-6548 Cloud Software Group NetScaler ADC CVSS 5.5 · 1 article · 2026-09-28
• CVE-2023-25608 Fortinet FortiAP-W2 CVSS 5.2 · 1 article · 2026-09-28
• CVE-2026-88772 Memory overflow vulnerability leading to Remote Code Execution or Denial of Service CVSS 9.5 · 3 articles · 2026-09-28
• CVE-2026-88771 A remote code execution vulnerability exists due to improper input validation, which can allow an unauthenticated attacker to execute arbitrary commands CVSS 9.5 · 3 articles · 2026-09-28