Claude Code agent allowed data exfiltration via DNS requests

Jun 2025

CVE-2025-55284 (CVSS 7.1) allowed attackers to bypass Claude Code's confirmation prompts and exfiltrate sensitive data from developers' computers through DNS requests. Prompt injection embedded in analyzed code could leverage auto-approved common utilities to silently steal secrets.

Incident Details

Perpetrator:AI coding agent

Severity:Facepalm

Blast Radius:Claude Code users on versions prior to 1.0.4 exposed to data exfiltration via prompt injection in code repositories

Tech Stack

Claude Code

References

Embrace The Red: Claude Code Data Exfiltration with DNS (CVE-2025-55284) ↗NVD: CVE-2025-55284 ↗Snyk: Command Injection in @anthropic-ai/claude-code ↗