careful

Scan shell and Git commands for destructive patterns before execution.

1|Updated Apr 26, 2026
One-click install
npx skills add https://github.com/tblakex01/conductor-playground --skill careful-tblakex01
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: careful
Source: https://github.com/tblakex01/conductor-playground/tree/main/.claude/skills/gstack/careful
Command: npx skills add https://github.com/tblakex01/conductor-playground --skill careful-tblakex01

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides safety guardrails for destructive commands, warning before actions like rm -rf, git reset --hard, kubectl delete, and similar operations to help prevent data loss. It also allows overriding warnings when necessary, making it suitable for careful production work.

Core Features & Use Cases

  • Pre-run command scanning that detects risky patterns (rm -rf, git reset --hard, kubectl delete, etc.)
  • Explicit user override to proceed after a warning, enabling controlled experimentation
  • Safe exemptions for certain build artifacts or benign operations to prevent false positives
  • Seamless integration via a PreToolUse hook that enforces guardrails in tooling

Quick Start

Enable careful mode and confirm before executing any destructive command.

Frequently Asked Questions about careful

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

FAQPage Schema
How do I add a safety prompt before running destructive shell commands like rm -rf?

To add a safety prompt before running destructive shell commands like rm -rf, you can use a pre-run hook that scans input commands for risky patterns and returns a permission decision requiring explicit user confirmation before execution.

Can I override the safety warnings for git reset --hard if I know the command is necessary?

Yes, you can override the safety warnings for git reset --hard. The precheck mechanism allows explicit user override to proceed after a warning, enabling controlled experimentation and careful production work when you intentionally execute destructive operations.

Does this command guardrail approach work for CI/CD pipelines and production environments?

Yes, this command guardrail approach works for CI/CD pipelines and production environments. The safety hook applies to developer workflows across local machines, CI/CD pipelines, and production touchpoints where destructive operations may be invoked to prevent data loss.

What destructive commands are detected by the pre-run hook during shell and Git actions?

The pre-run hook detects destructive commands during shell and Git actions by scanning for risky patterns such as rm -rf, git reset --hard, kubectl delete, and similar operations to prevent data loss before execution.

How do I prevent false positives when safety guardrails block benign build artifact deletions?

To prevent false positives when safety guardrails block benign build artifact deletions, the command scanning mechanism supports safe exemptions for certain build artifacts or benign operations, allowing them to execute without triggering the destructive command warning.

Why does a PreToolUse hook help prevent data loss in developer workflows?

A PreToolUse hook helps prevent data loss in developer workflows by enforcing guardrails directly in tooling, scanning input commands for risky patterns before they execute and requiring an explicit confirmation to proceed with destructive operations.