skills.rest
OTTTTTO
OTTTTTOCOMMUNITY
Explore all skills by this author →
GitHub

safe-exec

Intercept shell commands and require user approval for risky executions.

3|Updated Feb 1, 2026
Search Tags:#security#risk assessment#command execution#audit log#shell commands#ai agent safety
One-click install
npx skills add https://github.com/OTTTTTO/safe-exec --skill safe-exec
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: safe-exec
Source: https://github.com/OTTTTTO/safe-exec/tree/main
Command: npx skills add https://github.com/OTTTTTO/safe-exec --skill safe-exec

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires jq, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill acts as a crucial safety layer for AI Agents, preventing them from executing potentially destructive or insecure shell commands that could lead to data loss, system damage, or security breaches.

Core Features & Use Cases

  • Command Interception: Monitors and intercepts shell commands before execution.
  • Risk Assessment: Analyzes commands for dangerous patterns and assigns risk levels (CRITICAL, HIGH, MEDIUM, LOW).
  • Approval Workflow: Requires explicit user approval for risky commands via in-session notifications.
  • Audit Logging: Maintains a comprehensive log of all command executions, approvals, and rejections for traceability.
  • Use Case: When an AI Agent is asked to "delete all old log files," SafeExec will intercept the rm -rf command, notify the user of the HIGH risk, and wait for explicit approval before execution.

Quick Start

Enable SafeExec by telling your AI assistant to "Enable SafeExec".