Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
agentxagi avatar

Gstack Careful

  • 1 installs
  • 1 repo stars
  • Updated March 26, 2026
  • agentxagi/zero-inc

gstack-careful is a Claude Code skill that intercepts Bash commands with a hook and warns before running destructive operations.

About

This skill installs a safety guardrail that checks every Bash command against destructive patterns before it runs and asks for confirmation on a match. It protects against operations like rm -rf, DROP TABLE, git push --force, git reset --hard, and kubectl delete. A developer activates it when touching production, debugging live systems, or working in a shared environment.

  • Warns before destructive shell commands via a PreToolUse hook
  • Protects rm -rf, DROP TABLE, force-push, git reset --hard, kubectl delete and more
  • User can override each warning; safe exceptions like node_modules are allowed

Gstack Careful by the numbers

  • 1 all-time installs (skills.sh)
  • Ranked #1,835 of 2,203 Security skills by installs in the Skillselion catalog
  • Data as of Jul 7, 2026 (Skillselion catalog sync)
At a glance

gstack-careful capabilities & compatibility

Capabilities
command guardrail · destructive command check · safety hook
Use cases
security audit
From the docs

What gstack-careful says it does

Safety mode is now **active**. Every bash command will be checked for destructive patterns before running.
SKILL.md
These patterns are allowed without warning: - `rm -rf node_modules` / `.next` / `dist` / `__pycache__` / `.cache` / `build` / `.turbo` / `coverage`
SKILL.md
npx skills add https://github.com/agentxagi/zero-inc --skill gstack-careful

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs1
repo stars1
Last updatedMarch 26, 2026
Repositoryagentxagi/zero-inc

What it does

Guard an agent session against destructive shell commands by warning before rm -rf, DROP TABLE, force-push and similar operations.

Who is it for?

Sessions touching production, live systems, or shared environments where a bad command is costly.

Skip if: Routine local work where destructive-command risk is negligible.

When should I use this skill?

You are about to work on prod or a live system and want a confirmation gate on destructive commands.

What you get

A session-scoped hook that pauses and asks before any destructive command executes.

  • Session-scoped destructive-command confirmation hook

By the numbers

  • 8 destructive command patterns protected
  • 8 safe rm -rf exceptions allowed

Files

SKILL.mdMarkdownGitHub ↗

/gstack-careful — Destructive Command Guardrails

Safety mode is now active. Every bash command will be checked for destructive patterns before running. If a destructive command is detected, you'll be warned and can choose to proceed or cancel.

What's protected

PatternExampleRisk
rm -rf / rm -r / rm --recursiverm -rf /var/dataRecursive delete
DROP TABLE / DROP DATABASEDROP TABLE users;Data loss
TRUNCATETRUNCATE orders;Data loss
git push --force / -fgit push -f origin mainHistory rewrite
git reset --hardgit reset --hard HEAD~3Uncommitted work loss
git checkout . / git restore .git checkout .Uncommitted work loss
kubectl deletekubectl delete podProduction impact
docker rm -f / docker system prunedocker system prune -aContainer/image loss

Safe exceptions

These patterns are allowed without warning:

  • rm -rf node_modules / .next / dist / __pycache__ / .cache / build / .turbo / coverage

How it works

The hook reads the command from the tool input JSON, checks it against the patterns above, and returns permissionDecision: "ask" with a warning message if a match is found. You can always override the warning and proceed.

To deactivate, end the conversation or start a new one. Hooks are session-scoped.

Related skills

FAQ

How does it warn?

A PreToolUse Bash hook checks the command and returns permissionDecision "ask" with a warning when a destructive pattern matches.

Are any deletes allowed silently?

Yes - rm -rf on node_modules, .next, dist, __pycache__, .cache, build, .turbo and coverage are safe exceptions.

Securityappsec

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.