codex-hybrid-login

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs collecting an SSH password and explicitly tells the agent to prepend ssh commands with sshpass -p '<password>', which requires the LLM to embed the secret verbatim into generated commands (high exfiltration risk).

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (medium risk: 0.60). The skill directs the agent to modify remote configuration files, copy model directories, install/use sshpass, delete cache files and restart processes on the target server — operations that change system state and can require elevated privileges, so it poses a moderate risk even though it does not explicitly ask for sudo or account creation.