domains / families / security / safe-exec

safe-exec

separate

Safe command execution for OpenClaw Agents with automatic danger pattern detection, risk assessment, user approval workflow, and audit logging. Use when agents need to execute shell commands that may be dangerous (rm -rf, dd, fork bombs, system directory modifications) or require human oversight. Provides multi-level risk assessment (CRITICAL/HIGH/MEDIUM/LOW), in-session notifications, pending request management, and non-interactive environment support for agent automation.

When to use

• executing shell commands that may be dangerous
• commands requiring human oversight
• multi-level risk assessment
• in-session notifications
• pending request management
• non-interactive environment support for agent automation
• audit logging of operations

When not to use

• network calls
• external notifications
• background monitoring processes

Skill page

Source