High-Risk Operation Guard
Enforces explicit approval gates before executing destructive changes, production writes, secret modifications, or privilege escalations.
The High-Risk Operation Guard skill establishes strict safety boundaries for AI coding agents by blocking sensitive actions by default until explicit user approval is granted. It mandates affirmative consent prior to executing destructive file modifications, production database writes, credential or secret updates, and privilege level adjustments. By establishing that the lack of explicit approval is a definitive stop condition rather than an invitation to infer user intent, this skill safeguards development and production environments against unintended outages and unauthorized system changes.
Key Features
01Default-blocked state preventing automated execution of sensitive actions
02Zero-assumption policy treating absent approval as an absolute stop condition
03Strict approval gates for destructive file and environment modifications
04Safeguards against secret manipulation and credential exposure
050 GitHub stars
06Protection against unverified production writes and live deployments
Use Cases
01Enforcing compliance and security guardrails across enterprise developer workflows
02Guarding production environments and secret stores against unapproved autonomous agent writes
03Preventing accidental data loss and destructive system commands during automated refactoring