guard

Enforce directory-bound edits and warn before destructive commands.

1|Updated Mar 1, 2026

Search Tags:#security #bash #safety #freeze #guard #destructive #careful

Ready to run

npx skills add https://github.com/anishs1207/agentic-cli --skill guard-anishs1207

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: guard
Source: https://github.com/anishs1207/agentic-cli/tree/main/.agents/skills/gstack-guard
Command: npx skills add https://github.com/anishs1207/agentic-cli --skill guard-anishs1207

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Guards against destructive command execution and enforces directory-bound edits to protect critical systems during admin tasks and debugging.

Core Features & Use Cases

Combines careful-style destructive-command warnings with freeze-style edit boundaries to prevent accidental damage.
Provides pre-tool safety hooks that verify Bash actions, edits, and writes before they run.
Use Case: Enable guard mode during prod maintenance to ensure edits stay inside a safe directory and destructive commands are warned.

Quick Start

Activate guard mode and set a freeze directory to restrict edits to a safe path.

Frequently Asked Questions about guard

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

FAQPage Schema

How do I prevent destructive bash commands from running during production maintenance?▼

You can prevent destructive bash commands by enabling guard mode, which applies pre-tool safety hooks to verify actions and warns against destructive command execution before they run.

How do I enforce directory-bound edits to restrict file modifications to a safe path?▼

To enforce directory-bound edits, activate guard mode and set a freeze directory, which applies edit boundaries to ensure modifications stay restricted inside a designated safe path.

What is pre-tool safety checking for bash actions and file writes?▼

Pre-tool safety checking is a mechanism that verifies bash actions, edits, and writes against safety boundaries and destructive command patterns before they are executed.

Can I use safety hooks to protect systems when multiple contributors are debugging?▼

Yes, guard mode is specifically designed for admin, debugging, and production maintenance scenarios where multiple contributors could cause unintended changes to critical systems.

Does guard mode work with cross-skill hooks for comprehensive write safety?▼

Yes, guard mode integrates cross-skill hooks with careful and freeze components, combining destructive-command warnings with edit boundaries to provide comprehensive write safety.