Zed editor AI agent could bypass permissions for arbitrary code execution
Aug 2025
CVE-2025-55012 (CVSS 8.5) allowed Zed's AI agent to bypass user permission checks and create or modify project configuration files, enabling execution of arbitrary commands without explicit approval. Attackers could trigger this through compromised MCP servers, malicious repo files, or tricking users into fetching URLs with hidden instructions.
Incident Details
Perpetrator:AI coding agent
Severity:Facepalm
Blast Radius:All Zed users with Agent Panel prior to version 0.197.3
Tech Stack
Zed EditorAI Agent Panel