Detect destructive commands and prompt for confirmation before execution.
npx skills add https://github.com/drt0927/tw-overlay --skill careful-drt0927
Please help me install this Agent Skill. Skill: careful Source: https://github.com/drt0927/tw-overlay/tree/main/.gemini/skills/gstack-careful Command: npx skills add https://github.com/drt0927/tw-overlay --skill careful-drt0927
This skill provides safety guardrails for destructive commands by warning before execution and allowing users to override warnings when appropriate, reducing the risk of accidental data loss.
Enable careful mode to automatically warn before executing potentially destructive commands.
High-intent search queries and answers about installing and using this skill.
Yes, real-time detection of dangerous operations works across shells and tooling environments including Git, Kubernetes, and database contexts, using a pre-execution hook that can be extended to additional tools in the future.
Interactive safety prompts apply to Kubernetes and database contexts, detecting dangerous operations like kubectl delete and DROP TABLE to warn you before execution and allow a user-driven override flow.
The best way to guard destructive shell commands is implementing a pre-execution hook that detects dangerous operations and issues interactive warnings, allowing you to override the alert when an intentional force-push is required.
You can track potentially destructive actions by enabling a safety guardrail that provides audit logging, recording usage analytics every time a destructive command pattern triggers a warning or is executed.
reomgitMonitor and restrict destructive shell commands with bash and grep.
mulyovedWarns before executing risky Bash commands via a PreToolUse hook with per-command overrides and usage logging.
apeejay-educationWarn before executing destructive shell, Git, Kubernetes, and database commands.
reomgitMonitor and restrict directory edits and destructive commands in Bash environments.
