agent-collaboration

Standardize cross-agent collaboration with worktree awareness and shared decision recording.

Updated Apr 3, 2026
One-click install
npx skills add https://github.com/sachinnair90/gitgloss --skill agent-collaboration-sachinnair90
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-collaboration
Source: https://github.com/sachinnair90/gitgloss/tree/main/.copilot/skills/agent-collaboration
Command: npx skills add https://github.com/sachinnair90/gitgloss --skill agent-collaboration-sachinnair90

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It centralizes repeatable collaboration rules for multi-agent work so teams do not duplicate guidance across every charter and can coordinate consistently.

Core Features & Use Cases

  • Worktree Awareness: Uses the TEAM ROOT path as the authoritative repository reference so agents operate from the correct location.
  • Decision Recording: Captures decisions in a shared inbox drop-box so other agents can review and act on them without editing the central log directly.
  • Cross-Agent Communication: Guides agents to request input from teammates when a task depends on another agent’s domain or authority.
  • Reviewer Protocol: Clarifies how rejection feedback should assign revision ownership to keep review workflows unambiguous.
  • Use Case: A distributed coding squad can coordinate implementation, review, and follow-up changes without conflicting file ownership or misplaced assumptions about the repo root.

Quick Start

Use the agent-collaboration skill to help me follow the team workflow rules, record any decision that affects others, and flag when another agent should be brought in.

Frequently Asked Questions about agent-collaboration

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

FAQPage Schema
How do I coordinate worktree awareness across multiple AI agents?

Multi-agent worktree awareness is coordinated by designating a TEAM ROOT path as the authoritative repository reference. Agents use this shared path to operate from the correct location and maintain consistent repository awareness across implementation and review tasks.

What is the best way to record shared decisions in a multi-agent coding squad?

Shared decisions in multi-agent coding squads are recorded by writing to a shared inbox drop-box. Agents submit decisions to this inbox so teammates can review and act on them without directly editing the central decision or history files.

How does cross-agent communication work when an implementation task depends on another agent's domain?

Cross-agent communication prompts an agent to explicitly request input from teammates when a task depends on another agent's domain or authority. This ensures dependencies are flagged and resolved before implementation proceeds.

How do I manage revision ownership for code review rejections in a multi-agent workflow?

Revision ownership for code review rejections is managed through a reviewer protocol that assigns revision tasks directly. This protocol clarifies which agent is responsible for addressing rejection feedback to keep review workflows unambiguous.

Can I use agent collaboration rules for a distributed coding squad handling handoff scenarios?

Yes, agent collaboration rules apply to distributed coding squads managing implementation, review, and handoff scenarios. The standardized rules prevent conflicting file ownership and misplaced repo root assumptions across team members.

Why should agents avoid direct edits to shared decision logs in cross-agent coordination?

Agents should avoid direct edits to shared decision logs to prevent conflicting file ownership and data corruption. Centralized decision recording through the inbox drop-box ensures all teammates can review and act on changes safely.