SafeExec - Safe Command Execution Provides secure command execution capabilities for OpenClaw Agents with automatic interception of dangerous operations and approval workflow. Features - 🔍 Automatic danger pattern detection - Identifies risky commands before execution - 🚨 Risk-based interception - Multi-level assessment (CRITICAL/HIGH/MEDIUM/LOW) - 💬 In-session notifications - Real-time alerts in your current terminal/session - ✅ User approval workflow - Commands wait for explicit confirmation - 📊 Complete audit logging - Full traceability of all operations - 🤖 Agent-friendly - Non-inter…