Intercept destructive Bash commands and warn users before execution.
npx skills add https://github.com/mi179/claude-skills --skill careful-mi179
Please help me install this Agent Skill. Skill: careful Source: https://github.com/mi179/claude-skills/tree/main/user/gstack/careful Command: npx skills add https://github.com/mi179/claude-skills --skill careful-mi179
Safety mode provides guardrails that warn before destructive commands like rm -rf, git reset --hard, kubectl delete, and more, preventing accidental data loss in prod or shared environments.
Enable safety mode for destructive commands by running the pretool hook before executing risky operations.
High-intent search queries and answers about installing and using this skill.
You can intercept destructive bash commands by enabling a pre-execution safety check that detects patterns like rm -rf and warns you before execution. This guardrail prevents accidental data loss by returning a permission decision to proceed or cancel.
The best way to prevent accidental data loss during live debugging is applying a pre-execution hook that intercepts risky operations. It performs checks for destructive patterns and prompts a warning, allowing you to verify the command before proceeding.
Yes, you can get a warning before running git reset --hard or kubectl delete by using a safety hook. It scans for these specific destructive patterns and returns a permission decision, offering an option to proceed only after acknowledging the warning.
Yes, the destructive command safety check supports safe exceptions for build artifacts. While it intercepts patterns like rm -rf and git push --force by default, it allows defined safe operations to proceed without triggering the warning.
IchenDEVExecute arbitrary Bash shell commands with a 30-second timeout.
thanh-abaiiMonitor bash scripts for destructive commands and issue warnings.
NavarrePrattAutomate Kubernetes cluster management with kubectl and Helm workflows.
knopkiAnalyze shell scripts with ShellCheck and configure linting rules.
