Scans shell/system commands and requests confirmation before executing destructive ones.
npx skills add https://github.com/geoffreywoo/clawfable --skill careful-geoffreywoo
Please help me install this Agent Skill. Skill: careful Source: https://github.com/geoffreywoo/clawfable/tree/main/.claude/skills/gstack/careful Command: npx skills add https://github.com/geoffreywoo/clawfable --skill careful-geoffreywoo
This skill provides safety guardrails to prevent destructive commands from being executed without explicit confirmation, reducing the risk of data loss in development, CI, and production environments.
Enable careful mode and start issuing commands; the system will warn before executing any destructive operation.
High-intent search queries and answers about installing and using this skill.
You can prevent destructive bash commands by enabling a PreToolUse hook that scans instructions and prompts for confirmation before executing patterns like rm -rf. This guardrail intercepts the command and emits an 'ask' warning to stop accidental execution.
Yes, you can add a confirmation prompt before running git reset --hard in a CI pipeline by using a PreToolUse hook. It detects destructive commands and requires explicit confirmation to proceed, preventing unintended data loss during automated workflows.
A PreToolUse hook works by scanning your command input before execution. When it matches a destructive pattern, it emits a permissionDecision of 'ask' with a warning, pausing the action until you explicitly approve it.
Yes, safety guardrails exist for kubectl delete operations in production environments. They intercept these destructive commands, prompt for explicit confirmation, and log usage analytics for auditing to prevent catastrophic data loss.
Yes, you can whitelist safe command patterns while maintaining strict guardrails for others. This allows specific exceptions for trusted commands while still intercepting and warning before any unrecognized destructive operations.
Yes, the command warning hook logs usage analytics for auditing risky commands. It records when destructive patterns are matched and warnings are emitted, helping you analyze behavior and improve safety policies over time.
steadyfallIntercept destructive shell commands with a PreToolUse hook to warn before execution.
NightreaverExecute whitelisted shell commands on remote hosts with audit logging.
chicanoandres702Automate defensive Bash scripting with strict mode and safe argument parsing.
26mitch26Monitor Bash script execution for destructive commands and request confirmation.
