AnthropicGHSA-jh7p-qr78-84p7
Anthropic: information disclosure
Medium5.3CVE-2026-21852 · Published Jan 20, 2026
A vulnerability in Claude Code's project-load flow allowed malicious repositories to exfiltrate data including Anthropic API keys before users confirmed trust. If a user started Claude Code in an attacker-controller repository, and the repository included a settings file that set ANTHROPIC_BASE_URL to an attacker-controlled endpoint, Claude Code would issue API requests before showing the trust prompt, including 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 the latest version.
Affected versions
| Package | Affected | Fixed in |
|---|---|---|
| @anthropic-ai/claude-code npm | < v2.0.65 | v2.0.65 |
Details and references
More Anthropic advisories
All Anthropic| Date | Advisory | Severity | Fixed in |
|---|---|---|---|
| Feb 6 | Command Injection via Piped sed Command Bypasses File Write Restrictions | High7.7 | v2.0.55 |
| Feb 6 | Sandbox Escape via Persistent Configuration Injection in settings.json | High7.7 | v2.1.2 |
| Feb 6 | Permission Deny Bypass Through Symbolic Links | Low2.3 | v2.1.7 |
| Feb 3 | Command Injection in find Command Bypasses User Approval Prompt | High7.7 | v2.0.72 |
| Feb 3 | Path Restriction Bypass via ZSH Clobber Allows Arbitrary File Writes | High7.7 | v2.0.74 |
| Feb 3 | Domain Validation Bypass Allows Automatic Requests to Attacker-Controlled Domains | High7.1 | v1.0.111 |