skills.rest
agentmatters
agentmattersOFFICIAL
Explore all skills by this author →
WebsiteGitHub

safety-guard

Intercept destructive commands and restrict edits during production maintenance.

3|2|Updated Mar 8, 2026
Search Tags:#production#safety#guardrails#command-interception#mode-control#destructive-ops
One-click install
npx skills add https://github.com/agentmatters/mullai-bot --skill safety-guard-agentmatters
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: safety-guard
Source: https://github.com/agentmatters/mullai-bot/tree/main/src/Mullai.Skills/Skills/claude-code-everything/safety-guard
Command: npx skills add https://github.com/agentmatters/mullai-bot --skill safety-guard-agentmatters

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents destructive operations on production systems and during autonomous agent runs by enforcing safe work modes.

Core Features & Use Cases

  • Careful Mode intercepts destructive commands before execution and warns.
  • Freeze Mode locks edits to a specific directory tree to prevent unintended changes.
  • Guard Mode combines protection and restrictions for maximum safety.
  • Unlock command to revert protections when needed.
  • Easy integration with existing workflows and observability tooling.

Quick Start

Enable safety-guard and set it to Guard Mode for a specific project directory to prevent destructive actions.

Frequently Asked Questions about safety-guard

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent destructive commands from running on production systems?▼

To prevent destructive commands on production systems, enable safety-guard's Careful Mode, which intercepts potentially harmful commands before execution and issues a warning to halt unintended operations.

What is the best way to lock edits to a specific directory tree during sensitive migrations?▼

The best way to lock edits to a specific directory tree during sensitive migrations is using Freeze Mode, restricting modifications to an isolated path to prevent unintended changes to broader production code.

How do I stop an autonomous agent run from making unintended changes to my codebase?▼

To stop an autonomous agent run from making unintended changes, activate Guard Mode to combine command interception and directory restrictions, enforcing maximum safety across your autonomous workflows.

Can I revert safety protections when I need to execute destructive operations?▼

Yes, you can revert safety protections when you need to execute destructive operations by using the dedicated unlock command, which removes the active work mode restrictions.

Does safety-guard integrate with existing observability tooling and workflows?▼

Yes, safety-guard supports easy integration with existing workflows and observability tooling, allowing you to monitor command interception and mode control events alongside your current production maintenance processes.

Related Skills

Explore More Software Engineering Skills→
GGEduGGEdu

safety-guard

Enforce safety guards on agent actions to prevent destructive operations in production environments.

Community
Intermediate
ZESCODEZESCODE

safety-guard

Intercept and validate system commands and file operations against safety patterns.

Community
Advanced
ericdahl-devericdahl-dev

guard

Monitor Bash commands for destructive actions and restrict edits to a specified directory.

Official
Intermediate
yo-stevenyo-steven

nemoclaw-user-manage-policy

Manage NemoClaw sandbox egress by adding, removing, or modifying allowed network endpoints.

Community
Intermediate