safe-execlisted

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. Includes BashAST (skills/bash_ast/) — a Python AST parsing security layer implementing Claude Code's four-layer bash security model (AST parsing, regex validation, permission engine, OS sandbox).
Deep-pahoehoe114/openclaw-claude-code-integration · ★ 2 · AI & Automation · score 78

Install: claude install-skill Deep-pahoehoe114/openclaw-claude-code-integration

# 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-interactive mode support for automated workflows - 🔧 **Platform-agnostic** - Works independently of communication tools (webchat, Feishu, Telegram, etc.) - 🔐 **Security-focused** - No monitoring, no external notifications, no network calls ## Agent Mode When called by OpenClaw agents in non-interactive environments: - **Automatic bypass of confirmation prompts** - Prevents agent hanging - **Full audit logging** - All executions recorded with mode label (agent_auto vs user_approved) - **Safety preserved** - Danger pattern detection and risk assessment remain active - **Intended use case** - Automated workflows with human oversight via audit logs **Environment variables:** - `OPENCLAW_AGENT_CALL` - Set by OpenClaw when agent executes commands - `SAFE_EXEC_AUTO_CONFIRM` - Manual override to auto-approve LOW/MEDIUM risk commands **Security Note:** Agent mode