team-communication-protocols

Defines messaging protocols for multi-agent team coordination including plan approval and shutdown workflows.

Updated Apr 23, 2026
One-click install
npx skills add https://github.com/SanketAdlak/PDMProjectDesign --skill team-communication-protocols-sanketadlak
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: team-communication-protocols
Source: https://github.com/SanketAdlak/PDMProjectDesign/tree/main/.agents/skills/team-communication-protocols
Command: npx skills add https://github.com/SanketAdlak/PDMProjectDesign --skill team-communication-protocols-sanketadlak

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Coordinating multiple AI agent teammates often fails due to unclear messaging norms: wasted broadcasts, missed integration handoffs, deadlocks, and ungraceful shutdowns. This Skill provides structured protocols so agent teams communicate predictably and avoid common coordination anti-patterns. ## Core Features & Use Cases - Message Type Selection: Guidance on when to use direct messages, broadcasts, and shutdown requests, with JSON payload examples. - Plan Approval Workflow: A review loop where the team lead approves or rejects an implementer's plan before work begins. - Graceful Shutdown Protocol: A sequenced shutdown_request/shutdown_response flow with rejection handling and team cleanup. - Message Templates: Ready-to-use templates in references/messaging-patterns.md for task assignment, blocker reports, integration notifications, and review summaries. - Use Case: When spawning a new agent team, use this Skill to establish communication norms, route plan approvals through the lead, and shut down teammates cleanly after all tasks complete. ## Quick Start Use the team-communication-protocols skill to set up messaging norms and a plan approval workflow for my newly spawned agent team.

Frequently Asked Questions about team-communication-protocols

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

FAQPage Schema
How do I coordinate communication between multiple AI agents?

Use direct messages for point-to-point updates, reserve broadcasts for critical shared-resource changes, and route plan approvals through the team lead. Establish these norms when the team is spawned and use teammate names rather than IDs for all messaging.

When should I use broadcast vs direct message in agent teams?

Use direct messages by default for task updates, questions, and integration notifications. Reserve broadcasts for critical blockers or changes to shared resources, since each broadcast sends one message per teammate and scales poorly with team size.

How do I implement a plan approval workflow for AI agents?

Spawn the teammate with plan mode required so it explores read-only and calls ExitPlanMode, which sends a plan_approval_request to the lead. The lead replies with a plan_approval_response containing the request_id and an approve flag, optionally with feedback.

Why is my agent teammate not responding to messages?

Check the teammate's task status first. If idle, it likely finished its task and awaits new work or shutdown; if active, it will process messages after its current operation completes. Idle teammates should be assigned work or shut down.

How do I gracefully shut down a team of AI agents?

Send a shutdown_request to each teammate, wait for shutdown_response approvals, and retry after rejections once the teammate finishes its current work. After all teammates exit, call TeamDelete to remove team resources.

What causes deadlocks between AI agent teammates?

Deadlocks occur when two teammates each wait for the other's output before proceeding. The lead should break the cycle by sending one teammate a direct message with a stub or partial result so it can unblock and continue.