pair-agent

Coordinates multi-agent browser sharing sessions with isolated tabs and token-based security.

Updated Apr 11, 2026
One-click install
npx skills add https://github.com/akiyoshi/gstack-copilot-jp --skill pair-agent-akiyoshi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pair-agent
Source: https://github.com/akiyoshi/gstack-copilot-jp/tree/main/.github/skills/pair-agent
Command: npx skills add https://github.com/akiyoshi/gstack-copilot-jp --skill pair-agent-akiyoshi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Sharing a browser session between AI agents without duplicating work and coordinating actions.

Core Features & Use Cases

  • Multi-agent browser sharing via a safe, one-time setup key.
  • Parallel browsing across agents with isolated tabs per agent.
  • Secure session tokens and optional remote tunneling for cross-machine pairing.
  • Use Case: Engineers collaborate with another agent to inspect and navigate a site simultaneously.

Quick Start

Pair an additional AI agent to your browser and begin coordinated browsing sessions across agents.

Frequently Asked Questions about pair-agent

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

FAQPage Schema
How do I share a browser session between multiple AI agents?

To share a browser session between multiple AI agents, you pair them using a local HTTP browser server and a one-time setup key. This synchronizes their browsing tasks and isolates tabs per agent to prevent overlapping work.

How does browser sharing work for AI agents across remote machines?

Cross-machine browser sharing works by applying optional remote tunneling alongside secure session tokens. Agents connect to a local HTTP browser server using a one-time setup key, establishing synchronized browsing and session ownership management across local and remote environments.

Do I need a local HTTP server to pair AI agents for coordinated browsing?

Yes, you need a local HTTP browser server to pair AI agents. It acts as the central hub for the connection, utilizing a one-time setup key and token-based sessions to securely manage access and coordinate browsing tasks.

Can multiple AI agents browse in parallel without duplicating work?

Yes, paired AI agents can browse in parallel without duplicating work. The session provides isolated tabs per agent, ensuring that each agent navigates independently while sharing the same overall browser environment for synchronized task coordination.

What is the best way to manage session ownership for remote AI agents?

The best way to manage session ownership for remote AI agents is through token-based sessions. By utilizing a one-time setup key and optional remote tunneling, you can securely connect agents and control access across local and remote environments.