Vulnerabilidad · Publicado 12/10/2026
Network-AI from 5.14.0 before 5.15.2 contains a security control bypass vulnerability in the ClaudeHookBridge PreToolUse hook, which matches deny patterns only against the first present candidate field. A prompt-injected agent can place dangerous payloads in fields like Write content or an MCP script parameter beside a benign file_path or command, evading denyPatterns to write web shells or execut…
Network-AI from 5.14.0 before 5.15.2 contains a security control bypass vulnerability in the ClaudeHookBridge PreToolUse hook, which matches deny patterns only against the first present candidate field. A prompt-injected agent can place dangerous payloads in fields like Write content or an MCP script parameter beside a benign file_path or command, evading denyPatterns to write web shells or execute scripts.
Score: 7.4/10 — Severidad: HIGH — Estado NIST: Deferred
CVSS:3.1/AV:L/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H
CWE-436
Publicado en NIST NVD.