wizard

Require explicit human approval before executing destructive operations.

8|1|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/ek33450505/claude-agent-team --skill wizard-ek33450505
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wizard
Source: https://github.com/ek33450505/claude-agent-team/tree/main/skills/wizard
Command: npx skills add https://github.com/ek33450505/claude-agent-team --skill wizard-ek33450505

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents accidental irreversible changes by requiring explicit human approval before any destructive operation, reducing the risk of data loss, broken deployments, or lost history.

Core Features & Use Cases

  • Step-by-step approval gates: Announce intent, show impact, provide rollback steps, wait for explicit approval, then execute and confirm.
  • Scope coverage: Applies to file system operations, git history rewrites and force-pushes, database schema or mass-data changes, dependency removal, and infrastructure/CI alterations.
  • Safety-first workflow: Groups related small changes into a single gate, avoids batching unrelated destructive actions, and halts on failures for operator review.
  • Use case: Safely perform a production database migration by enumerating affected tables and rows, presenting rollback commands, and requiring an authorized "approved" confirmation before proceeding.

Quick Start

Activate wizard mode and require explicit approval before performing any destructive operation.

Frequently Asked Questions about wizard

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

FAQPage Schema
How do I require human approval before destructive git pushes or database migrations?

To require human approval before destructive git pushes or database migrations, enforce step-by-step gates that announce intent, list impact, provide rollback instructions, and wait for explicit consent before executing. This prevents accidental irreversible changes and data loss.

What is a human-in-the-loop approval gate for destructive operations?

A human-in-the-loop approval gate for destructive operations is a safety mechanism that halts automated workflows to present impact details and rollback steps, capturing explicit operator consent before executing irreversible actions like file deletions or infrastructure updates.

How do I safely execute a production database migration with rollback instructions?

To safely execute a production database migration, enumerate affected tables and rows, present rollback commands, and require an authorized approval confirmation before proceeding. The workflow halts on failures for operator review and provides immediate post-action confirmations.

Does the approval gate workflow apply to force git pushes and infrastructure updates?

Yes, the approval gate workflow applies to force git pushes, infrastructure updates, file system operations, dependency removal, and database schema changes, grouping related small changes into a single gate while avoiding batching unrelated destructive actions.

When should I not use automated batching for destructive file system operations?

You should not use automated batching for destructive file system operations when the actions are unrelated, as the safety-first workflow groups related small changes into a single gate but explicitly avoids batching unrelated destructive actions to ensure proper operator review.