guide-assistant

Guide users through AI ops assistant conversation, slash commands, and skill workflows.

4|Updated May 9, 2026
One-click install
npx skills add https://github.com/liaoxinjie666/protocol-proxy --skill guide-assistant-liaoxinjie666
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guide-assistant
Source: https://github.com/liaoxinjie666/protocol-proxy/tree/main/skills/system/guide-assistant
Command: npx skills add https://github.com/liaoxinjie666/protocol-proxy --skill guide-assistant-liaoxinjie666

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps users quickly understand how to operate the AI operations (智控助手) assistant, manage conversations and skills, and use slash commands and shortcuts effectively.

Core Features & Use Cases

  • Conversation & context management: Create, switch, and compress chat history to control context usage and token limits.
  • Client slash commands: Use browser-side commands like help, clear, new, compact, and model switching, with clear command priority behavior.
  • Skill system guidance: Trigger skills manually, allow AI auto-triggering via natural-language triggers, and create/edit/delete custom skills.
  • Advanced task execution concepts: Understand how delegation to sub-agents and tasks works, including roles, states, and safety limits.
  • Operational governance: Learn how tool approval and agent/persona (SOUL) concepts influence behavior and safety.

Quick Start

Ask the assistant: “Explain how to use 智控助手, including slash commands, starting a new conversation, managing skills, and how to interrupt a response.”

Frequently Asked Questions about guide-assistant

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

FAQPage Schema
How do I manage AI assistant conversations and control token limits?

You can manage AI assistant conversations by creating, switching, and compressing chat history to control context usage and token limits. This helps maintain efficient dialogue flow without exceeding processing constraints.

What slash commands are available for browser-side AI assistant operations?

Available client slash commands include help, clear, new, compact, and model switching. These commands execute browser-side operations with clear priority behavior to manage the chat interface.

How does the skill system work for triggering tasks in an AI ops assistant?

The skill system allows manual triggering or AI auto-triggering via natural language. Users can create, edit, and delete custom skills to manage specific operational task workflows.

Can I delegate tasks to sub-agents within the AI operations assistant?

Yes, task delegation to sub-agents is supported. The system defines specific roles, states, and safety limits to ensure sub-agents execute delegated tasks securely and effectively.

How do tool approval and agent personas influence AI assistant safety?

Tool approval and agent personas (SOUL) govern operational behavior and safety. They define how the AI assistant interacts with tools and ensure actions align with predefined safety protocols.

What is the best way to interrupt an AI assistant response during task execution?

You can interrupt an AI assistant response using built-in interruption controls. The skill system defines specific trigger mechanics and interruption controls to halt ongoing workflows safely.