guard-users-claude

Enforce policy-based guardrails and approvals for Claude CLI workflows.

Updated Feb 23, 2026

Search Tags:#cli #claude #classification #policy #approval #guardrails #risk

Ready to run

npx skills add https://github.com/AndrewRober/CodexCli-GuardRails --skill guard-users-claude

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: guard-users-claude
Source: https://github.com/AndrewRober/CodexCli-GuardRails/tree/main/skills/guard-users-claude
Command: npx skills add https://github.com/AndrewRober/CodexCli-GuardRails --skill guard-users-claude

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Refuse catastrophic actions, enforce scoped approvals, and minimize secret leakage in Claude CLI workflows.

Core Features & Use Cases

Prevent root/system-destruction and dangerous operations through policy enforcement.
Require explicit approvals for high-risk actions and provide clear risk notes.
Apply across workspace-scoped Claude CLI sessions to ensure consistent safety.

Quick Start

Configure a policy payload for a destructive command with a scoped, explicit approval requirement.

Frequently Asked Questions about guard-users-claude

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

FAQPage Schema

How do I enforce approval guardrails for Claude CLI workflows?▼

Approval guardrails for Claude CLI workflows are enforced by applying policy-based classifications to operations. This requires explicit approvals for destructive, credentialed, and high-risk actions within a workspace to prevent catastrophic outcomes.

How do I prevent destructive operations in Claude CLI sessions?▼

Destructive operations in Claude CLI sessions are prevented through strict policy enforcement and deterministic risk classification. The system refuses root and system-destruction commands by requiring scoped, explicit approval payloads before execution.

Can I scope Claude CLI approvals to specific workspace operations?▼

Yes, you can scope Claude CLI approvals to specific workspace operations. Policy payloads configured with YAML frontmatter metadata apply targeted guardrails to destructive or credentialed actions within that specific workspace context.

How do I minimize secret leakage when using Claude CLI?▼

Secret leakage is minimized by enforcing workspace-scoped guardrails and strict payload formats. Policy classifications intercept high-risk and credentialed operations, requiring explicit approvals before Claude CLI can process them.

Does Claude CLI risk classification require specific payload formats?▼

Yes, Claude CLI risk classification requires strict payload formats and YAML frontmatter metadata for name and description. This ensures deterministic risk classification outputs for policy enforcement across all operations.